Secondary Logo

Journal Logo

Global Health: Original Clinical Research Report

Overcoming Language Barriers Using an Information Video on Spinal Anesthesia for Cesarean Delivery: Implementation and Impact on Maternal Anxiety

Purcell-Jones, Jessica M. A. MBBS, BSc, DA*; Haasbroek, Marlis MBChB, BSc, DA, PG Dip Pain; Van der Westhuizen, Justine L. MBChB, DA*; Dyer, Robert A. FCA (SA), PhD*; Lombard, Carl J. MSc, PhD; Duys, Rowan A. MBChB, MRCP, MMed (Anaes)*

Author Information
doi: 10.1213/ANE.0000000000004243

Abstract

KEY POINTS

  • Question: In the context of a doctor–patient language barrier, does a spinal anesthesia information video narrated in Xhosa reduce anxiety in Xhosa women about to have spinal anesthesia for cesarean delivery?
  • Findings: In patients with higher baseline anxiety scores, a greater decrease in anxiety scores was observed after a spinal anesthesia information video than after a “usual care” explanation.
  • Meaning: Supplying video material with the narration conducted in the patient’s “mother tongue” is associated with lower anxiety scores in anxious patients, in the context of a doctor–patient language barrier.

South Africa is a multilingual country with 11 official languages. Patients and doctors infrequently share the same first language. In the Western Cape province, most doctors’ first language is English, then Afrikaans, with a small number of Xhosa speakers. Conversely, state hospital patients speak Xhosa (40%–50%), Afrikaans (25%–30%), and English (20%–30%).1 In addition to language difficulties, educational and cultural disparities frequently occur between doctors and their Xhosa patients.2,3 Medical translation should be undertaken by trained professional interpreters.4 However, few official translators are used in South African public hospitals, resulting in reliance on translation by staff, relatives, and other patients.1 This unofficial translation has been shown to be of poor quality due to misinterpretation, omissions, time pressure, and resentment.4,5 Such poor communication has been associated with patient dissatisfaction and anxiety about the potential negative impact on their treatment.1,4

The cesarean delivery rate in South Africa is increasing,6 and guidelines recommend regional anesthesia in the absence of contraindications.7 Spinal anesthesia explanations can be challenging in the light of poor correlation between doctor’s and patient’s first language in the public sector.

Anesthesia information videos can supplement doctors’ explanations and have been shown to reduce anxiety but have mostly been ineffective in improving patient satisfaction.8 An appropriate information video may bridge both language and cultural differences, with the potential to reduce patient anxiety and increase satisfaction. Smartphones can be used as a display medium for information videos, and with the almost complete penetration of cellular telephone ownership among health care workers, even in developing countries,9 such videos may be used worldwide to overcome disparities in doctor–patient communication.

This pilot study implemented and assessed the acceptability of a spinal anesthesia information video narrated in Xhosa, before elective cesarean delivery. The effect of the video on patient anxiety and satisfaction was assessed in first language Xhosa-speaking patients. We compared anxiety before and after our usual prespinal anesthesia explanation in a control group and before and after the spinal anesthesia information video in an intervention group. Finally, we assessed maternal satisfaction and asked women whether they would recommend the video to other patients. We aim to use the baseline data from this study to inform sample size calculations for future studies.

The primary hypothesis was that exposure to a Xhosa-narrated spinal anesthesia information video before elective cesarean delivery would be associated with lower anxiety levels in Xhosa patients, when compared with the “usual care” verbal prespinal explanation. The secondary hypothesis was that an improvement in patient satisfaction scores would be observed.

METHODS

This article adheres to Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) guidelines. After approval from the Human Research Ethics Committee of the Faculty of Health Sciences of the University of Cape Town, first language Xhosa-speaking women, over the age of 18 years, scheduled for an elective cesarean delivery, were invited to participate in this pilot study. All patients enrolled in the study signed a consent form written in Xhosa and English, the day before surgery. Illiterate patients had the form read out to them. Patients were enrolled sequentially at 2 secondary level hospitals in Cape Town, New Somerset Hospital and Mowbray Maternity Hospital, on the days on which the study investigators were assigned to the participating hospitals.

This study was not blinded and consisted of 2 sequential phases of recruitment. Phase 1 patients received the “usual care” verbal explanations of their spinal anesthesia (control group). The anesthesiologists, who were not part of the investigating team, were asked to give their usual explanation, either using a Xhosa-speaking staff member as an interpreter, when available, or in English. Phase 2 patients watched the Xhosa-narrated video explanation of their spinal anesthesia on the smartphone of their anesthesiologist (intervention group); phase 2 commenced after completion of phase 1. Patients in each group received individual explanations relating to the spinal anesthesia immediately before entering the operating theater for their cesarean delivery. Anesthesiologists participating in the study only viewed the video and the English translation of the Xhosa narrative when phase 2 had commenced, eliminating the risk of modification of their “usual care” explanation in phase 1. They were informed that the video did not serve as a substitute for the informed consent process.

The video was 3 minutes long and followed the journey of a Xhosa woman having a spinal anesthetic for her cesarean delivery. The majority of the video is narrated by a Xhosa-speaking nurse. Part of the video consists of a description by the patient of her experience of the spinal anesthesia (Supplemental Digital Content 1–2, Video, https://links.lww.com/AA/C827, Document 1, https://links.lww.com/AA/C828). The patient consented to participation in the video 24 hours before the recording session and gave written consent for the inclusion of the video as a supplement in any publication arising from the study, as well as for continued use of the video in the hospital setting.

The primary outcome was the anxiety score after a “usual care” verbal prespinal anesthesia explanation in a control group and after our spinal anesthesia information video in an intervention group. The null hypothesis was that there would be no difference in anxiety scores postexplanation between the intervention and control groups. Anxiety was assessed using a Numerical Visual Analog Anxiety Scale (NVAAS) (Supplemental Digital Content 3, Document 2, https://links.lww.com/AA/C829). The NVAAS was given to the patient in the preoperative room by the anesthesiologist, who was not part of the investigating team. If the patient was illiterate, it was read out verbatim by Xhosa-speaking nursing staff. The NVAAS is a validated score for anxiety assessment and shows positive correlation with other more complex anxiety scores.10 The simplicity of a numerical rating scale allows rapid assessment of anxiety and lends itself to easy translation, particularly in a population of low medical literacy.11 The minimum clinically important difference (MCID) on a 100-mm visual analog scale for anxiety is represented by a 15-mm change.12 This would correlate with a 1.5-point change on our 10-point NVAAS. Thus, a 1.5-point difference in the mean postexplanation NVAAS scores between the intervention and control groups was used to represent clinical significance in this study. Anxiety assessments were made immediately pre- and postspinal anesthesia explanation.

The secondary outcome was patient satisfaction, which was compared between the intervention and control groups. This was assessed using the Maternal Satisfaction Scale for Cesarean Section (MSSCS), the day after spinal anesthesia. The MSSCS was given to the patient by the study investigators to complete in their own time. If the patient was illiterate, it was read out verbatim by Xhosa-speaking nursing staff. The MSSCS is a validated scale consisting of 22 items, each rated on a 7-point scale.13 A recent study added 2 items asking patients to rate their ability to ask questions about the procedure and whether they received enough anesthesia information.14 These items were included in this study, bringing the total number of items to 24. In the study hospitals, mothers only hold their babies once the operation is complete, and not on the operating table. To reflect this protocol, the order of the items on the MSSCS was adjusted (Supplemental Digital Content 4, Document 3, https://links.lww.com/AA/C830). One of the items asks whether partner interaction was satisfactory. Because many Xhosa women do not have their partners in theater for the birth, this item did not apply in some cases. For mothers without a partner present, the questionnaire was reduced to 23 items, and all questionnaire results were expressed as a percentage to allow comparison between 23- and 24-item questionnaires.

Baseline data included age, indication for cesarean delivery, the language in which the control group explanation was given, self-reported English fluency of the patient, and baseline NVAAS scores prespinal anesthesia explanation.

It was specified in the protocol that, in the event of conversion to general anesthesia, or a maternal or fetal critical incident requiring admission to the intensive care unit (ICU), the patient was excluded from further postoperative assessments. If any of the data sets were incomplete, the collected data in such cases were retained for analysis.

The COnsensus based Standards for the selection of health Measurement INstruments (COSMIN) cultural validity checklist was used to ensure that the video translation was conducted at the appropriate cultural and educational level.15 Two officially trained medical translators and 5 Xhosa-speaking nursing staff were independently involved in translating the documents and video script, checking phraseology until a consensus agreement was reached as to the wording. The translation team was briefed to screen for educational and cultural appropriateness. The video was narrated by a Xhosa-speaking nurse and reviewed by 3 Xhosa nursing staff to ensure clear and comprehensible audio commentary. The translated consent, NVAAS and MSSCS, received a trial run on 5 Xhosa patients who met the study criteria to ensure ease of understanding.

Statistical Methods

Descriptive statistics (mean, standard deviation [SD], and proportion) were calculated by group, for age, indication for cesarean delivery, English literacy, NVAAS score pre- and postexplanation, and MSSCS score. To confirm the comparability of the groups at baseline, the NVAAS prescore was compared using an ordinal logistic regression, age was compared using a 2-sample t test, indication for cesarean delivery using the Pearson χ2 test, and patient-reported fluency for English using the Fisher exact test. Standardized differences between intervention and control were calculated for baseline covariates to facilitate baseline comparisons. The stddiff program from Stata was used for these calculations and is based on the approach of Yang and Dalton16 which is applicable to both continuous and categorical data. An unadjusted analysis of between-group difference in mean change in NVAAS score was performed using a 2-sample t test assuming equal variances. To extend this comparison to a model that adjusted for baseline covariates, as well as provide an adjusted effect estimate of the video intervention, and test for the significance of the association between the intervention and the postexplanation NVAAS score, a linear regression model was used. An adjustment was made for potential baseline confounding variables because this was a nonrandomized study. The regression model included baseline covariates (preexplanation NVAAS score, age, and English fluency), and an indicator for the intervention, as well as an interaction term between the intervention indicator and preexplanation NVAAS score. From this model, we estimated the mean difference between the groups in postexplanation NVAAS score for each level of the preexplanation NVAAS score, with 95% confidence intervals (CIs), and for the multiple significance testing (K = 11), we adjusted the significance criterion using the Bonferroni correction (.05/11 =.0045). The selection of covariates was based on clinical relevance (preexplanation NVAAS score and age) and important baseline differences between the participants in the 2 phases of recruitment (English fluency). As a sensitivity analysis, an ordinal regression model of postexplanation NVAAS score was done using the same fixed-effects model under the assumption of proportional odds.

Given the skewed MSSCS scores, a main effects only (no interactions) quantile regression model was used to compare the median satisfaction score between the intervention and control groups adjusting for baseline covariates, age and NVAAS preexplanation score and English fluency. A between-group comparison was performed, using the χ2 test, of the patient-rated opportunity to ask questions, and patient assessment of the adequacy of the information they received about their procedure. Statistical analysis was performed using Stata 15.1 (StataCorp, College Station, TX).

A sample size calculation for between-group comparison of postspinal explanation NVAAS was not performed because baseline NVAAS scores are not well documented in our population. For this pilot study, a convenience sample size was chosen of ≥75 patients per group. It was estimated that this would allow for adequate capacity to complete recruitment within the 12-month study period.

RESULTS

Over a 1-year period, 176 patients were eligible for the study when the investigators were present, and 1 declined to participate. One hundred seventy-five women were recruited to the study: 92 patients in the control and 83 in the intervention group. No patients withdrew from the study after recruitment and consent. The primary outcome was recorded in all recruited patients. With respect to the secondary outcomes, the postoperative MSSCS data were omitted in 15 patients in the control group and 4 in the intervention group (Figure 1).

Figure 1.
Figure 1.:
Study flow diagram. aTen incomplete MSSCS forms, 3 neonatal intensive care unit admissions, 1 postpartum hemorrhage, 1 conversion to general anesthesia. bFour neonatal intensive care unit admissions. MSSCS indicates Maternal Satisfaction Scale for Cesarean Section; NVAAS, Numerical Visual Analog Anxiety Scale.

Descriptive statistics for age, pre- and postexplanation NVAAS scores, maternal satisfaction scores, and indication for cesarean delivery for the 2 groups are reported in Table 1. There were no significant between-group differences in mean patient age (P = .439; standardized difference = −0.11), mean preexplanation NVAAS score (P = .652; standardized difference = 0.07), or indication for cesarean delivery (P = .154, standardized difference = 0.06). All patients’ first language was Xhosa. The patient self-reported English fluency was significantly different between the 2 groups (P = .005; standardized difference = 0.05), with the intervention group reporting greater English fluency (Table 1). In the control group, 86% of spinal anesthesia explanations were given in English, 2% by a Xhosa interpreter, and 12% in English combined with rudimentary Xhosa.

Table 1.
Table 1.:
Descriptive Statistics
Figure 2.
Figure 2.:
Box and whisker plot of NVAAS score pre- and postspinal explanation. NVAAS indicates Numerical Visual Analog Anxiety Scale.

Figure 2 shows a box and whisker plot of pre- and postexplanation NVAAS scores in the intervention and control groups. An unadjusted analysis showed a mean (SD) decrease in NVAAS score (mean pre- minus postexplanation NVAAS score [Table 1]) of 1.6 (3.5) and 0.7 (2.3), in the intervention and control groups, respectively, and this difference of 0.9 points (95% CI, 0.02–1.8 points) was statistically significant (P = .046).

Primary Result

There was a significant interaction between the intervention indicator and preexplanation NVAAS score (P = .002) in the adjusted analysis of postexplanation NVAAS score (Figure 3). Patients with preexplanation NVAAS scores ≤4 in the intervention group showed no difference in postexplanation NVAAS scores from the same subset in the control group, whereas those with preexplanation NVAAS scores ≥5 showed a significant intervention effect using the adjusted significance table (Table 2). Forty-six percent of patients in the intervention (39/83) and the control (42/93) groups had preexplanation NVAAS scores of ≥5. The estimated intervention effect for a participant with severe anxiety (preexplanation NVAAS score = 10 points) was 2.9 points (95% CI, 1.4–4.4 points; P < .001). The sensitivity analysis for the postexplanation NVAAS score (ordinal regression model) confirmed the interaction between the preexplanation NVAAS score and the intervention indicator, and the same prescore threshold for an intervention effect.

Table 2.
Table 2.:
Estimated Effect of Video Conditional on the Preexplanation NVAAS Score
Figure 3.
Figure 3.:
A linear regression model for postexplanation NVAAS score, showing the influence of preexplanation NVAAS score on the difference between the intervention and control groups. Individual scores are also represented. Solid shapes indicate estimated mean values with 95% confidence intervals and open shapes individual data. NVAAS indicates Numerical Visual Analog Anxiety Scale.

There was no significant difference in the MSSCS scores between the 2 groups when adjusted for age, English fluency, and preexplanation NVAAS score, median difference between intervention and control groups −0.53 (95% CI, −3.4 to 2.4; P = .717) (Table 3). Eighty-three of 84 patients in the intervention group would recommend the video to other patients.

Table 3.
Table 3.:
Quantile Regression Model for Evaluating the Association Between the MSSCS Score and the Intervention, Adjusted for Covariates

Patient-rated adequacy of information supplied, and of the opportunity to ask questions, showed no between-group difference. Eighty-seven percent and 86% of the intervention group patients versus 83% and 82% of the control group, respectively, “strongly agreed” with the statements “I received enough information about my anaesthetic,” and “I was able to ask questions if I was unsure about my operation” (P = .306 and P = .609, respectively).

DISCUSSION

This pilot study investigated the use of spinal anesthesia information videos narrated in Xhosa, for Xhosa-speaking patients before elective cesarean delivery. Our study shows that in patients with higher preexplanation NVAAS scores, the information video is associated with lower postexplanation NVAAS scores than the usual care explanation. In patients with low baseline anxiety scores, there appeared to be no additional benefit. No difference was observed in patient satisfaction, whether a verbal or video spinal explanation was used. Our patient population endorsed the intervention, with the majority recommending that the video be shown to other patients. Our results also suggest that patients felt free to ask questions after the video. In addition, we have described the novel use of smartphones as a display medium for patient information videos and for utilizing videos to bridge the doctor–patient language barrier. This type of intervention is thus feasible and potentially scalable to all areas where health care workers carry smartphones.

A 2003 systematic review examining the impact of exposure to media-based anesthesia explanations8 showed an improvement in patient anxiety but no difference in patient satisfaction. In the only previous study investigating the use of a spinal anesthesia information video before elective cesarean delivery, no impact on anxiety or satisfaction was shown.14 However, unlike in the present study, their analysis did not adjust for baseline anxiety score and this could account for the differences in outcome. Our study also showed no impact of the video on maternal satisfaction. It could be hypothesized that having safely delivered a healthy baby, maternal satisfaction would be high when assessed the day after the procedure.

Patients were recruited at 2 hospital sites over a 12-month period (6 months for each group). We specified a pragmatic minimum sample size of 75 per group. The numbers recruited during the specified time period exceeded the minimum sample size. An a priori sample size calculation was not done due to the paucity of data describing baseline variability in NVAAS scores in our population. The current pilot study data provide the necessary basis for such a sample size calculation (Table 1).

A NVAAS was chosen for the study over the traditional visual analog scale because it was felt that the concept was easier to understand for our patient population, with their low degree of literacy. No study has investigated the MCID for NVAAS. A 15-mm change is the MCID for traditional visual analog scales for anxiety. This would equate to a 1.5-point change on a NVAAS and was used as the MCID between groups in this study. Studies show a high correlation between visual analog scales and numerical visual analog scales for the assessment of pain.17,18 There have been no such comparisons for anxiety scores, and the assumption was made that a similar correlation exists.

The video production cost R12,000 (approximately $900), including editing and studio recording time. Therefore, considering the potential benefits, it is relatively inexpensive and feasible in all languages. To reduce cost further, 1 film could be used as the basis for multiple superimpositions of various languages, enhancing affordability in lower- and middle-income countries.

A study limitation was that language fluency was self-reported; however, objective measures are challenging in this situation. Most patients reported some level of English fluency, and most explanations in the control arm were given in English. However, it has been shown that both patients and staff who have identified themselves as being reasonably fluent in a second language are not always as proficient as they believe, and the communication disparity between doctor and patients may be underestimated.3 In our experience, patients with self-reported English fluency display minimal fluency with English medical terminology. Correct interpretation of a patient’s understanding of the explanation can be limited by cultural reticence in Xhosa patients to ask questions because they may feel that this indicates disrespect or are fearful of the unbalanced power dynamics in the doctor–patient relationship.19 When doctors perceive adequate fluency in English, they may be less likely to check the patient’s level of understanding.3

Patients were not randomly assigned to the control or intervention group. Instead the study was run in 2 sequential phases. This may have influenced the results because the theater environment (eg, stress levels in theater and theater team dynamics) could have influenced the patients’ anxiety or satisfaction. A sequential approach was chosen for the study to avoid bias, because there was concern that once the anesthesiologists had viewed the video script, they may alter their “usual care” spinal anesthesia explanation. The limitations of our before–after design include confounding by time, possible regression to the mean, the lack of concurrent controls, and the Hawthorne effect. If the latter effect had any influence, it might have reduced the impact of the intervention because anesthesiologists might have been more thorough in their “usual care” explanation.

Although the mean postexplanation NVAAS scores were statistically significantly lower in the intervention group if preexplanation scores were ≥5, none of the lower bounds of the 95% CI of the difference between means in this range was higher than the prespecified clinically significant MCID of 1.5 points (Table 2). This may reflect inadequate statistical power for this outcome in this pilot study.

Due to simple omission by patients, 10 of the MSSCS scores were incomplete in the control group. This was unlikely to have had any statistical impact because the analysis showed consistently high satisfaction scores.

The MSSCS scores and NVAAS have not been validated in Xhosa, only in English. The COSMIN taxonomy checklist was used to optimize translation of these scores and ensure cultural validity.15

CONCLUSIONS

This study investigated the use of a spinal anesthesia information video in the context of a language and cultural barrier. It also introduced a novel use of cellular telephones to distribute such information videos. In patients with higher levels of anxiety prespinal anesthesia, the video intervention was associated with a greater reduction in anxiety scores than the usual verbal explanation. There was a high degree of acceptance, which suggests that these videos can be used to ameliorate challenges posed by language barriers. This study will inform future work examining patient anxiety by providing baseline variability in anxiety scores, as well as an indication of effect size, to guide sample size calculations for NVAAS studies in similar populations. Home language information videos may be particularly valuable in groups of patients with high anxiety scores, for example, the setting of emergency cesarean delivery, in South Africa and worldwide.

ACKNOWLEDGMENTS

The authors thank their anesthesia colleagues at the University of Cape Town for aiding in data collection and Professor Bruce Biccard for aiding in statistical analysis of the data.

DISCLOSURES

Name: Jessica M. A. Purcell-Jones, MBBS, BSc, DA.

Contribution: This author helped conceive the idea for the study, produce the video, collect the data, and write the manuscript. She was the primary author.

Name: Marlis Haasbroek, MBChB, BSc, DA, PG Dip Pain.

Contribution: This author helped identify appropriate patients for the study at Mowbray Maternity Hospital, enroll patients, obtain consent, and complete the maternal satisfaction survey with the study patients.

Name: Justine L. Van der Westhuizen, MBChB, DA.

Contribution: This author helped identify appropriate patients for the study at New Somerset hospital, enroll patients, obtain consent, and complete the maternal satisfaction survey with the study patients.

Name: Robert A. Dyer, FCA (SA), PhD.

Contribution: This author helped supervise the primary author and compose and edit the manuscript.

Name: Carl J. Lombard, MSc, PhD.

Contribution: This author helped perform statistical analysis and compose and edit the manuscript.

Name: Rowan A. Duys, MBChB, MRCP, MMed (Anaes).

Contribution: This author helped supervise the primary author, produce the video for this study, and edit the manuscript.

This manuscript was handled by: Angela Enright, MB, FRCPC.

REFERENCES

1. Levin ME. Language as a barrier to care for Xhosa-speaking patients at a South African paediatric teaching hospital. S Afr Med J. 2006;96:1076–1079.
2. Lagerwerf L, Boer H, Wasserman H. Lagerwerf L, Boer H, Wasserman H. In: Health Communication in Southern Africa: Engaging With Social and Cultural Diversity. 2009:Amsterdam: Rozenberg publishers; 155–157.
3. Schyve PM. Language differences as a barrier to quality and safety in health care: the joint commission perspective. J Gen Intern Med. 2007;22suppl 2360–361.
4. Flores G. The impact of medical interpreter services on the quality of health care: a systematic review. Med Care Res Rev. 2005;62:255–299.
5. Saohatse MC. Communication problems in multilingual speech communities. South African J African Lang. 1998;18:111–117.
6. Farina Z, Rout C. “But it’s just a spinal”: Combating increasing rates of maternal death related to spinal anaesthesia. S Afr Med J. 2013;103:81–82.
7. Caesarean Section. 2011. National Institute for Health and Care ExcellenceAvailable at: https://Www.Nice.Org.Uk/Guidance/Cg132. Accessed May 17, 2019.
8. Lee A, Chui PT, Gin T. Educating patients about anesthesia: a systematic review of randomized controlled trials of media-based interventions. Anesth Analg. 2003;96:1424–1431.
9. Zurovac D, Otieno G, Kigen S, et al. Ownership and use of mobile phones among health workers, caregivers of sick children and adult patients in Kenya: cross-sectional national survey. Global Health. 2013;9:20.
10. Elkins G, Staniunas R, Rajab MH, Marcus J, Snyder T. Use of a numeric visual analog anxiety scale among patients undergoing colorectal surgery. Clin Nurs Res. 2004;13:237–244.
11. Hawker GA, Mian S, Kendzerska T, French M. Measures of adult pain: Visual Analog Scale for Pain (VAS Pain), Numeric Rating Scale for Pain (NRS Pain), McGill Pain Questionnaire (MPQ), Short-Form McGill Pain Questionnaire (SF-MPQ), Chronic Pain Grade Scale (CPGS), Short Form-36 Bodily Pain Scale (SF. Arthritis Care Res. 2011;63:240–252.
12. Williams VS, Morlock RJ, Feltner D. Psychometric evaluation of a visual analog scale for the assessment of anxiety. Health Qual Life Outcomes. 2010;8:57.
13. Morgan PJ, Halpern S, Lo J. The development of a maternal satisfaction scale for caesarean section. Int J Obstet Anesth. 1999;8:165–170.
14. Eley VA, Searles T, Donovan K, Walters E. Effect of an anaesthesia information video on preoperative maternal anxiety and postoperative satisfaction in elective caesarean section: a prospective randomised trial. Anaesth Intensive Care. 2013;41:774–781.
15. Mokkink LB, Terwee CB, Patrick DL, et al. The COSMIN checklist for assessing the methodological quality of studies on measurement properties of health status measurement instruments: an international Delphi study. Qual Life Res. 2010;19:539–549.
16. Yang D, Dalton JE. A unified approach to measuring the effect size between two groups using SAS®. SAS Glob Forum 2012. 2012:1–6.
17. Downie WW, Leatham PA, Rhind VM, Wright V, Branco JA, Anderson JA. Studies with pain rating scales. Ann Rheum Dis. 1978;37:378–381.
18. Bijur PE, Latimer CT, Gallagher EJ. Validation of a verbally administered numerical rating scale of acute pain for use in the emergency department. Acad Emerg Med. 2003;10:390–392.
19. Schlemmer A, Mash B. The effects of a language barrier in a South African district hospital. S Afr Med J. 2006;96:1084–1087.

Supplemental Digital Content

Copyright © 2019 International Anesthesia Research Society