Demographic Factors Affect Scoliosis Research Society-22 Performance in Healthy Adolescents

         The Scoliosis Research Society (SRS) Instrument was developed by Haher et al in 1999 to assess postoperative patient outcome in adolescents with idiopathic scoliosis (AIS).1 This 24-item questionnaire assesses different domains of health related quality of life (HRQOL) including pain, general self-image, general function, level of activity, postoperative function, and satisfaction. This disease specific questionnaire was accepted as a simple and practical means of obtaining a patient’s perception of his/her disease. Any questionnaire measuring outcome must identify what is intended (validity), be reproducible, show internal consistency (reliability), and be sensitive to change over time.1–18 Outcome surveys must also have discriminant validity, or the ability to distinguish between persons without the disease and patients with varying disease severity.2 In the original version of the SRS instrument (SRS24), the validity and reliability were reported to be satisfactory, but the sensitivity to change over time was not clearly determined. Since this time, the SRS-24 has undergone a number of modifications to improve reliability and validity.3–5 White et al reported several shortcomings in the validity and reliability of the SRS questionnaire and its unknown responsiveness to change.6 Since that time another version of the SRS instrument was developed to include a mental health domain, comparing its validity with the Short Form-36 (SF-36). Usage of this version prompted two additional modifications to improve internal consistency, resulting in the current version, known as the SRS-22.3 Disease-specific questionnaires such as the SRS-22 are useful in measuring a patient’s HRQOL affected by a disease process.

          In a study conducted in an adult population comparing those with and without scoliosis, the discriminate validity of the SRS instrument was found to be excellent.2 However, for the adolescent population with regard to AIS, no comparative baseline exists for persons unaffected by this condition. Moreover, there is no literature describing how demographic factors may influence the SRS-22. Demographic factors such as gender, race and socioeconomic status can contribute to the variability within a group, which may pose a challenge to the interpretation of patients’ responses to the questionnaire.7 The purpose of this study is to establish a normative baseline useful for evaluating the discriminate validity of the SRS-22 in primary adolescent idiopathic scoliosis and to analyze the influence of demographic factors including race, gender, household status, and From the Department of Orthopaedic Surgery, NYU Hospital for Joint Diseases, New York, NY. Acknowledgment date: September 3, 2009. Revision date: October 19, 2009. Acceptance date: October 19, 2009. The manuscript submitted does not contain information about medical device(s)/drug(s). No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Address correspondence and reprint requests to Baron Lonner, MD, 820 2nd Ave, Suite 7A, New York, NY 10017; E-mail: BLonner@ nyc.rr.com 2134 household income, on individual domain and overall SRS-22 performance. Materials and Methods Following Institutional Review Board (IRB) approval, 450 healthy unaffected adolescents completed the SRS-22 from January to April 2009 (268F, 163 M, 19 unknown; mean age 16.0 years; range, 10–21).

         Parental consent was not required by the IRB. Individual domain and overall SRS-22 performance were assessed in all healthy adolescents excluding 2 questions (21 and 22) regarding postoperative satisfaction. The SRS-22 domain scores were calculated in the usual manner as described elsewhere.3 In addition, demographic questions were asked to estimate yearly household (less than $15,000, $15,000– $30,000; $30,000–$50,000, $50,000–$75,000, $75,000– $125,000, $125,000–$200,000, and greater than $200,000), race (white, Hispanic, African-American, Native American, Pacific Islander, Asian, or other), gender, weight, height, and household status (single vs. dual parent) (Figure 1). Students completing the anonymous survey were also given information regarding the SRS-22, study goals, confidentiality of responses, and additional contact information. Subjects Subjects age 10 to 21 years without significant medical illness or a prior history of scoliosis were asked to complete the questionnaires. Three hundred SRS-22 and demographic questionnaires were given to the principals of 2 New York public high schools located in Manhattan (males and females) and Staten Island (females only). The schools were located in separate boroughs of Manhattan with different student body profiles in terms of socioeconomic status and ethnicity. The principals of the schools were asked to give the SRS-22 and demographic questions to students at the school’s health clinic or during physical education class. Both principals were informed of the IRB approval, but were given the option to attain parental consent if deemed necessary by the local school board.

        At the Manhattan school, the students were asked to complete the survey during physical education class and 266 of 300 students completed the surveys. At the Staten Island school, parental consent was obtained prior after which 73 students completed the SRS-22 during physical education class. In addition, a total of 200 surveys were given to a major municipal hospital adolescent health clinic and 3 locally referring pediatrician offices. Physicians in the health clinic and in the private offices were asked to offer the SRS-22 to healthy adolescents without significant medical illness or a history of scoliosis. At total of 119 questionnaires were completed. Overall, 800 questionnaires were dispensed to local high schools and health clinics in the Manhattan area with a 56% response rate. 8 surveys were eliminated from the analysis due to age ineligibility and poor penmanship.

          Statistics After a description of the demographic data (global and by group), correlation between demographic data (age, gender, body mass index, race, household status, and income range) and clinical outcomes (SRS-22 domains) was completed using a Spearman coefficient. Subjects were then stratified by race, income and household status and differences among groups in terms of SRS-22 scores were evaluated using a 1-way ANOVA and a post hoc test. Demographic Questionnaire Household Status (please check the appropriate box) Single Parent Household  Dual Parent Household  Parent/Parents Combined Income Range (please check the appropriate box) $0- $15,000  $75,000- $125,000  $15,000- $30,000  $125,000-$200,000  $30,000- $50,000  $200,000 +  $50,000- $75,000  Racial Category Definition of Category  Asian Please Specify_______________ A person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent including: China, Japan, Korea, Malaysia, Pakistan, Thailand, Vietnam, Cambodia, India, and the Philippine Islands.  Black or African American Please Specify________________  Hispanic or Latino Please Specify________________ A person having origins in any of the black racial groups of Africa. A person of Cuban, Mexican, Puerto Rican, South/Central American, and other Spanish lt i i  Native American or Alaska Native Please Specify________________ culture or origin. A person having origins in any of the original peoples of North, Central, or South America, and who maintain tribal affiliations or community attachment Native Hawaiian or Pacific Islander Please Specify_________________  White or Caucasian community attachment. A person having origins in any of the original peoples of Hawaii, Guam, Samoa, or other Pacific Islands. A person having origins in any of the original Please Specify________________  Other (Please Specify) ______________ p gg y g peoples of Europe, the Middle East, or North Africa. Figure 1. Healthy adolescent demographic questionnaire used to household status, household income, and race. Adolescents without scoliosis were asked to complete these demographic questions in addition to the SRS-22. Demographics Affect SRS-22 in Healthy Adolescents • Verma et al 2135 Results Descriptive Statistics The healthy adolescent profile was as follows: 62% female and 38% male, average age 16 (range 9.3 to 21.8) years, average height 65 (range 41–86) inches, average weight 138 (range 69 to 294) pounds, with an average body mass index (BMI) of 22.8 (range 13.5 to 47.5) (Table 1).

          There was an 85.6% to 96.9% response rate in questions relating to these parameters. The largest racial group was of Hispanic origin, constituting of 198 (49%) of the subjects. Other racial groups consisted of 96 (24%) white, 66 (16%) African American, 20 (5%) Asian, with 26 (6%) adolescents being of another ethnicity or of mixed race. Overall there was a 90.3% response rate of the demographic question pertaining to race. Single and dual-parent homes respectively constituted 42.5% and 57.5% of the healthy adolescent households, with an 80% response rate to this question. Average household income was between 30,000– 50,000 dollars/y (income range 3) and 50,000 to 75,000 dollars/y (income range 4). The question regarding household income was answered by 59% of adolescents completing the SRS-22 questionnaire. Correlation Analysis Small correlations (all P
0.005, Table 2) were found between demographic data and clinical scores: income range and ethnicity were correlated with higher activity score (r 0.282 and r 0.217, respectively); while gender was correlated with mental (r 0.264) and mean (r 0.187) scores. As for the demographic parameters alone, the income range was moderately correlated with the household status (r 0.33) and the ethnicity (r 0.4). Analysis by Gender and Household In terms of gender and household status, healthy males tended to perform better than healthy females in the mental health domain (P
0.0001; 4.1 vs. 3.7) and in overall SRS-22 performance (P
0.0001; 4.2 vs. 3.7) (Table 3).

          Adolescents with dual parent households had higher activity scores (P
0.005; 4.2 vs. 4.0) and higher overall SRS-22 performance (P
0.005) compared to adolescents with single households. The difference in mean SRS-22 score between these groups, however, was small (4.2 vs. 4.1). Analysis by Race When stratifying the subjects by race, the one way ANOVA analysis did not reveal any significant differences in terms of Image, Mental and Mean SRS-22 scores (Figure 2). However, white adolescents tended to perform better in the SRS-22 activity domain as compared to Hispanic and other ethnicities (P
0.05; 4.3 vs. 4.0). African American adolescents also scored higher in the Table 1. Healthy Adolescent Demographics N 450 268 F, 163 M 19 no response (4.2%) Household status 153 single, 207 dual 92 no response (20%) Race 198 Hispanic 96 White 66 Black 20 Asian 26 other/mix 44 no response (9.7%) Income (range, 1–7) 3.6 (1–7) 185 no response (41%) Age (yr) 16 (10–22) 14 no response (3.1%) Height (inches) 65.0 (41–86) 28 no response (6.2%) Weight (lbs) 139.2 (69–294) 65 no response (14.4%) BMI 22.9 (13.5–47.5) Income: (1)
$15,000; (2) $15,000 –$30,000; (3) $30,000 –$50,000; (4) $50,000 –$75,000; (5) $75,000 –$125,000; (6) $125,000 –$200,000; (7) $200,000. Income, age, height, weight, and BMI given as “mean (range).” Table 2. Correlational Analysis R P Demographic parameter SRS domain Household status Activity 0.153
0.01 Dual households 3 more activity Income range Activity 0.282
0.0001 Larger income 3 more activity Racial Activity 0.217
0.0001 Gender Pain 0.103
0.05 Males 3 higher pain scores Income range Image 0.183
0.01 Income 3 better image score Gender Mental 0.264
0.0001 Males 3 higher mental domain scores Gender Mean 0.187
0.0001 Males 3 higher overall SRS scores Household status Mean 0.115
0.05 Dual households 3 higher overall score Income range Mean 0.163
0.01 Income 3 higher overall score Parameter Household income range 0.334
0.0001 Dual households 3 higher income Racial income range 0.4
0.0001 SRS indicates Scoliosis Research Society.

         Table 3. SRS-22 Varies With Gender and Household Status Parameter SRS Domain n Mean SD P Gender Female Mental 268 3.7 0.76 P
0.0001 Male 163 4.1 0.67 Female Mean 268 4 0.45 P
0.0001 Male 163 4.2 0.43 Household Single Activity 153 4.0 0.55 P
0.005 Dual 207 4.2 0.5 Single Mean 153 4.1 0.4 P
0.05 Dual 207 4.2 0.6 SD indicates standard deviation; SRS, Scoliosis Research Society. 2136 Spine • Volume 35 • Number 24 • 2010 pain domain than Hispanic adolescents (P
0.05; 4.5 vs. 4.3). Analysis by Income From the lowest income range up to 125,000 dollars/yr, household income had a positive effect on the activity domain (P
0.05; range, 3.8–4.3), image domain (P
0.05; range, 4.0–4.4), and mean (P
0.05; range, 4.0– 4.3) SRS-22 performance (Figure 3). There was one exception; incomes from 50,000 to 75,000 dollars/yr scored similarly to the lowest income bracket in the activity domain. Incomes above 75,000 to 125,000 dollars/yr did not have an additional effect on SRS-22 scores. Discussion Cultural differences can affect patients’ interpretation of the questions on any HRQOL questionnaire, as evidenced by validity studies performed on translated versions of the SRS-22 questionnaire.8–13 In addition to linguistic translations, however, cultural adaptations are also necessary to identify conceptual differences of the disease. Watanabe et al conducted a cross-cultural comparison of the SRS-24 and found Japanese outperformed American patients in pain (4.3 vs. 3.7), function (4.2 vs. 3.9), and activity domains (4.9 vs. 4.5).

         However, Japanese patients scored worse than Americans in the selfimage domain (3.5 vs. 4.0).12 In a previous study, a population of 72 healthy Japanese adolescents also performed worse on questions 14 and 15 (self image domain) than patients with mild scoliosis.13 The authors concluded that cultural differences between Japanese and American patients likely accounted for unexpected finding and noted few studies reporting normative data in adolescents.13 Alanay et al also reported satisfactory 4.4 *° * 4.2 4.0 SRS Activity Domain Score 4.6 4.4 4.2 SRS Pain Domain Score * * * ° * White Black Hispanic Other Ethnicity 3.8 White Black Hispanic Other Ethnicity 4.0 Activity Domain (p < 0.05): White vs Hispanic; White vs Other Image domain (p < 0.05): Black vs Hispanic y *° * Figure 2. Race was significantly associated with the activity and pain domains of the SRS-22. White adolescents showed higher activity scores compared to Hispanic and other ethnicities (P
0.05). African American adolescents showed higher pain scores than Hispanic adolescents (P
0.05). 4.6 4.4 4.2 SRS Activity Domain Score 4.6 4.4 4.2 SRS Image Domain Score * * *° ‡ ° † ° 1 2 3 4 5 6 7 4.0 3.8 3.6 1 2 3 4 5 6 7 4.0 3.8 *†‡° Income: 1) <$15K; 2) $15-30K; 3) $30-50K; 4) $50-75K; 5) $75-125K; 6) $125-200K; 7) >$200K Activity Domain: 1 vs 3 (p < 0.001); 1 vs 5 (p < 0.001); 1 vs 6 (p < 0.01); 1 vs 7 (p < 0.05) Image domain (p < 0.05): 1 vs 4; 1 vs 5 Mean SRS-22 (not shown): 1 vs. 5 ( p < 0.0 1 2 3 4 5 6 7 Income 1 2 3 4 5 6 7 Income *° *† ‡ ° Figure 3.

         Income was significantly associated with the SRS-22 activity and pain domain scores. Adolescents whose household income was less than 15,000 per year had significantly lower activity scores compared to adolescents with higher parent incomes. Similarly, adolescents with household incomes less than $15,000 per year had significantly lower image scores than adolescents with higher incomes (P
0.05). Demographics Affect SRS-22 in Healthy Adolescents • Verma et al 2137 internal consistency and only moderate validity in many questions using the Turkish version of the SRS-22. It is reasonable to consider, therefore, that differences in race within American adolescents may similarly influence SRS-22 performance. We find that race, gender, household income, and household status lead to significant differences on SRS-22 performance. Specifically, white race was associated with higher activity domain scores compared to Hispanic and “Other” ethnicities (4.3 vs. 4.0). African American adolescents also tended to score higher than Hispanic adolescents on the pain domain (4.5 vs. 4.3). Increased household income had a positive effect on activity (4.3 vs. 3.8), image (4.4 vs. 4.0), and mean SRS-22 score (4.3 vs. 4.0); however, incomes above 125,000 dollars/y did not have an added effect. Male adolescents tended to score higher on the mental health domain (4.1 vs. 3.7) and overall SRS-22 (4.2 vs. 4.0) compared to females adolescents, while dual parent adolescents tended to score higher on the activity domain (4.2 vs. 4.0) and mean SRS-22 compared to single parent adolescents (4.2 vs. 4.1). To our knowledge this is the first study to characterize the influence of demographic factors on the SRS-22 and provide the largest normative adolescent baseline for comparison to AIS.

         While the differences observed with race, income, and household status in this study are numerically small, they represent a clinically significant variation in SRS-22 performance. Prior studies have validated the SRS-22 citing similar differences in domain scores between comparison groups.1–5,14,16–18 Asher et al assessed the discriminate validity of the SRS-22 with a normal population of 19 adolescents, 68 patients with non surgical scoliosis, and 32 patients with preoperative scoliosis. The authors reported a 0.3, 0.5, 0.9, 0.5, and 0.6 variation in activity, pain, self-image, mental health, and overall SRS-22 scores respectively between the three groups. With the exception of the activity domain, the authors concluded that these differences in domain scores were demonstrative of good discriminative validity.17 For 18 patients with curves smaller than 20°; Cheung et al reported patients treated with bracing scored lower in the activity (4.49 vs. 4.91), pain (4.42 vs. 4.82), and self-image domains (3.62 vs. 4.31) than untreated patients with similar curves magnitudes. This effect size was found to represent a significant impact on the quality of life of the patient.18

          Similarly, Haher et al found that 26 healthy adolescents scored higher than scoliosis patients in all domains with the smallest affect size being 0.5 for the self image domain. The authors noted that comparisons between healthy adolescents and scoliosis patients were not matched for age and gender, which would have strengthened the comparison.1 In an age and gender matched comparison of healthy adults with and without scoliosis, Baldus et al reported significantly lower SRS scores in all domains for adult scoliosis versus adults without scoliosis. The smallest significant difference reported was between 4.1 and 4.3 in mental health domain. Overall, eight out of thirty reported differences in domain scores were within the 0.2 to 0.7 effect range.2 While the study by Baldus et al matched for age and gender, demographic factors were not considered. Noteworthy, the dataset in this study has proportionally more males than typically seen in the scoliosis population. Also, the percentage of African American and Hispanic patients may be larger than seen in some geographical areas. Lastly, the average income was between 30,000 and 75,000 dollars/yr which may also vary with region. Physicians using this normative dataset should consider the demographic profile of patients seen in their practice.

         While the SRS-22 is a simple tool for assessing outcomes in patients with scoliosis, demographic factors were found to influence the SRS-22 in adolescents without scoliosis. The effect of race, gender, income, and household status on the SRS-22 represents a meaningful clinical variation that is comparable to the influence of curve magnitude and bracing as reported by previous authors.1–5,14,16–18 Future studies using the SRS-22 for comparisons of healthy adolescents with AIS may consider matching patients based on age, gender, race, household income, and household status. Key Points ● We established a normative baseline for the SRS-22 in healthy adolescents. ● Individual demographics affect individual domain and overall SRS-22 scores. ● Race, income, gender, and single versus dual parent households affected the SRS-22 with a meaningful clinical difference.

References

1. Haher TR, Gorup JM, Shin TM, et al. Results of the Scoliosis Research Society Instrument for evaluation of surgical outcome in adolescent idiopathic scoliosis: a multicenter study of 244 patients. Spine 1999;24: 1435–40.

2. Baldus C, Bridwell KH, Harrast J, et al. Age-gender matched comparison of SRS instrument scores between adult deformity and normal adults. Are all SRS domains disease specific? Spine 2008;33:2214–8.

3. Asher MA, Lai SM, Burton D, et al. The reliability and concurrent validity of the Scoliosis Research Society-22 patient questionnaire for idiopathic scoliosis. Spine 2003;28:63–9.

4. Asher MA, Lai SM, Burton D. Further development and validation of the Scoliosis Research Society (SRS) outcomes instrument. Spine 2000;25: 2381–6.

5. Asher MA, Lai SM, Glattes C, et al. Refinement of the SRS-22 health-related quality of life questionnaire function domain. Spine 2006;31:593–7.

6. White SF, Asher MA, Lai SM, et al. Patients’ perceptions of overall function, pain, and appearance after primary posterior instrumentation and fusion for idiopathic scoliosis. Spine 1999;24:1693–1700.

7. Haher TR, Valdevit A. The use of outcomes instruments in the assessment of patients with idiopathic scoliosis. Instr Course Lect 2005;54:543–50.

8. Niemeyer T, Schubert C, Halm HF, et al. Validity and reliability of an adapted German version of Scoliosis Research Society-22 questionnaire. Spine 2009;34:818–21.

9. Cheung KM, Senkoylu A, Alanay A, et al. Reliability and concurrent validity of the adapted Chinese version of Scoliosis Research Society-22 (SRS-22) questionnaire. Spine 2007;32:1141–5.

10. Alanay A, Cil A, Berk H, et al. Reliability and validity of adapted Turkish 2138 Spine • Volume 35 • Number 24 • 2010 version of Scoliosis Research Society-22 (SRS-22) questionnaire. Spine 2005; 30:2464–8.

11. Climent JM, Bago J, Ey A, et al. Validity of the Spanish version of the Scoliosis Research Society-22 (SRS-22) patient questionnaire. Spine 2005; 30:705–9.

12. Watanabe K, Lenke L, Bridwell KH, et al. Cross-cultural comparison of the Scoliosis Research Society outcomes instrument between American and Japanese idiopathic scoliosis patients. Are there differences? Spine 2007;32:2711–4.

13. Watanabe K, Hasegawa K, Hirano T, et al. Use of the Scoliosis Research Society outcomes instrument to evaluate patient outcome in untreated idiopathic scoliosis patients in Japan. Part I: comparison with nonscoliosis group: preliminary/limited review in a Japanese population. Spine 2005;30: 1197–201.

14. Asher MA, Lai SM, Burton D, et al. Scoliosis Research Society-22 patient questionnaire. Spine 2003;28:70–3.

15. Lai SM, Asher M, Burton D. Estimating SRS-22 quality of life measures with SR-36: application in idiopathic scoliosis. Spine 2006;31:473–8.

16. Asher M, Lai SM, Burton D, et al. The influence of spine and trunk deformity on preoperative idiopathic scoliosis patients’ health-related quality of life questionnaire responses. Spine 2004;29:861–8.

17. Asher MA, Lai SM, Burton D, et al. Discrimination validity of the scoliosis research society-22 patient questionnaire: relationship to idiopathic scoliosis curve pattern and curve size. Spine 2003;28:74–8.

18. Cheung KM, Cheng EY, Chan SC, et al. Outcome assessment of bracing in adolescent idiopathic scoliosis by the use of the SRS-22 questionnaire. Int Orthop 2007;31:507–11.