Validation of the Spanish version of the Side Effect and Life Satisfaction Inventory in patients with epilepsy

Please download to get full document.

View again

of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Information Report

News & Politics


Views: 71 | Pages: 6

Extension: PDF | Download: 0

Validation of the Spanish version of the Side Effect and Life Satisfaction Inventory in patients with epilepsy
  Validation of the Spanish version of the Side Effect and Life Satisfaction Inventory inpatients with epilepsy Vicente Villanueva a, * , Antonio Gil-Nágel b , Eloy Elices c , José M. Serratosa d , Juan C. Sánchez-Alvarez e ,Mar Carreño f  , Javier Salas-Puig g , Joana Porcel h a Epilepsy Unit, N eurology Department, University Hospital La Fe de Valencia, Avenida Campanar, 21, Valencia 46009, Spain b Epilepsy Program, Neurology Department, Hospital Ruber International, Madrid, Spain c Neurology Department, Clinica Rotger, Palma de Mallorca, Spain d Epilepsy Unit, Neurology Department, Jiménez Díaz Foundation, Madrid, Spain e Neurology Department, Hospital San Cecilio, Granada, Spain f  Epilepsy Unit, Neurology Department, Clinic i Provincial Hospital, Barcelona, Spain g Neurology Department, University Hospital, Oviedo, Spain h Biometria Clínica S.L. Barcelona, Barcelona, Spain a r t i c l e i n f o  Article history: Received 30 July 2008Revised 3 September 2008Accepted 5 September 2008Available online 11 October 2008 Keywords: EpilepsyPsychometric validationHealth-related quality of lifeSide Effects and Life Satisfaction Inventory a b s t r a c t Objective:  Thegoal of thestudy describedhere was toobtain psychometric validationof theSpanishver-sion of the 38-item Side Effects and Life Satisfaction (SEALS) Inventory. Methods:  A cross-cultural adaptation of the inventory was performed. A total of 595 patients with epi-lepsywereincludedinamulticenter cross-sectional study. TheSEALSInventorywascompleted, togetherwith the Hospital Anxiety and Depression Scale and SF-12 Health Survey. Results:  The mean SEALS score was 60.7. SEALS presented high internal consistency, with a Cronbach  a coefficient of 0.93, and good test–retest reliability, with an intraclass correlation coefficient of 0.92.The patternof correlations withtheHospital AnxietyandDepressionScale andSF-12HealthSurveyindi-catedgoodconvergent anddivergent validity. SEALS scores discriminatedpatientsaccording toepilepsy-related factors, emotional disturbances, and the generic quality of life. Conclusion:  The Spanish version of the SEALS Inventory is a valid psychometric instrument. It may beused in routine clinical practice and in clinical trials in patients with epilepsy to capture the cognitiveand behavioral aspects of quality of life.   2008 Elsevier Inc. All rights reserved. 1. Introduction The prevalence of epilepsy in the Spanish population rangesfrom 4.21/1000 to 5.72/1000 in adults and those of school age,respectively [1,2]. In addition to good seizure control, improve- ment in quality of life (QoL) is considered a mandatory treatmentobjectiveinpatientswithepilepsy,supportedbythefactthatthesepatients report not only symptoms related to seizures, but also as-pects of their cognitive, social, and emotional functioning. In thiscontext, QoL questionnaires are a useful tool for developing opti-mal protocols for seizure management and also for clinical trialsof new drugs.A number of QoL questionnaires have been developed to assesshealth-relatedQoLinepilepsy. Thefirst epilepsy-specificquestion-naire published was the Washington Psychosocial Seizure Inven-tory (WPSI) [3]. This questionnaire was followed by the Walton Hospital Seizure Severity Scale [4], the Epilepsy Surgery Inventory(ESI-55) [5], and, finally, various versions of the Quality of Life inEpilepsy (QoLIE) scale [6–8], two of which have been validated in Spanish: the QoLIE-31 [9] and the QoLIE-10 [10]. Morerecently,amodifiedversionoftheSideEffectandLifeSat-isfaction Inventory (SEALS) has been proposed. The SEALS Inven-tory, designed to measure the effects of anticonvulsant therapyon psychosocial functioning (affect and cognition), was developedin the United Kingdom as a 50-item, self-report questionnaire, de-rived from the symptoms and epilepsy treatment side effects re-ported by a patient population [11] and was later refined to a38-item inventory [12,13]. Kugoh used the 50-item version of  SEALS in combination with the WPSI, and concluded that whenadministered together, both scales provided a good estimation of QoLinpatientswithepilepsy[14]The38-itemversionoftheSEALShas also been validated by Gillhamet al. [13] and was divided intofive subscales: termed Cognition, Dysphoria, Temper, Tiredness,and Worry. Since then, a large international program of cross-cul-tural adaptation of the SEALS Inventory has been undertaken; theFrench version of this scale has already been reported [15]. 1525-5050/$ - see front matter   2008 Elsevier Inc. All rights reserved.doi:10.1016/j.yebeh.2008.09.003* Corresponding author. Fax: +34 96 1973290. E-mail address: (V. Villanueva).Epilepsy & Behavior 14 (2009) 96–101 Contents lists available at ScienceDirect Epilepsy & Behavior journal homepage:  This study therefore presents the psychometric validation of theSpanish version of the 38-item Side Effect and Life SatisfactionInventory. 2. Methods Sixty-seven neurologists were invited to participate in a cross-sectional study conducted in Spain. The patients attended outpa-tient clinics located in 58 hospitals (listed in Appendix I—see Sup-plementary Material). The participant neurologists were asked toconsecutively recruit the first 10 patients seen at the outpatientclinic who met the following inclusion criteria: age greater than18 years, diagnosis of epilepsy for at least 1 year, treatment withstable antiepileptic drug therapy for 1 month prior to enrollment,and ability to give informed consent for the study. Patients withother concomitant conditions likely to affect cognition and/orQoL were excluded, as were those patients who were included inanother clinical study. Demographic data, medical history, andclinical characteristics of the epilepsy were documented by theinvestigator; the subjects were asked to complete the SEALS Inven-tory,aswellas the Spanish versionsof the SF-12HealthSurvey(SF-12) [16] and the Hospital Anxiety and Depression Scale (HADS)[17]. The SF-12is ageneric measureof QoL that comprises 12 itemsand provides two summary scores, Physical Component and Men-tal Component, and eight indices: General Health, Physical Func-tioning, Role Physical, Role Emotional, Bodily Pain, MentalHealth, Vitality, and Social Functioning. Higher scores on the SF-12 indicate better QoL. The HADS is a screening scale for emotional(anxiety and depression) symptoms widely used in the generalmedical setting. It consists of 14 items divided into two subscales:Anxiety (odd items) and Depression (even items). Patients are clas-sified according to their total score on each subscale into four cat-egories: scores <8 indicate the absence of symptoms, scores in therange 8–0 indicate a borderline or mild disorder, scores of 11–14indicate a moderate disorder, and scores >14 indicate the presenceof severe symptoms.The neurologist also recorded any adverse effect related to anti-epileptic drug (AED) treatment reported by the patient. Patienttreatment compliance during the previous week was also recorded.Additionally, the first two patients on stable medication re-cruited by each neurologist were asked to complete the SEALSInventory 2 weeks (±3 days) after inclusion.The translation and cross-cultural adaptation of the 38-itemEnglish version of SEALS proposed by Gillham et al. [12] were car-ried out first, as outlined in Fig. 1, to obtain a Spanish version con-ceptually equivalent to the srcinal English version and adapted tothe culture in which it was going to be used. The validated Spanishversion of the SEALS Inventory is provided in Appendix II.The study was approved by the University Hospital La Fe(Valencia) ethics committee. All patients gave written informedconsent.Data analysis was performed using the SPSS 15.0 statisticalpackage. Individual scale scores, as well as the overall score, werecalculated for those patients who had answered all the items on agiven individual scale and all the items on the SEALS.The mean, SD,median, and range were calculatedfor each individual scale and forthe overall scale. Ceiling and floor effects and data on question-naire feasibility were also obtained. An exploratory factor analysiswith varimax rotation was used to identify the questionnairestructure. For multitrait analysis, internal consistency of the ques-tionnaire and its scales was analyzed with Cronbach’s a coefficient.Reproducibility was assessed by test–retest using intraclass corre-lation between the same questionnaires completed by the samesubject at an interval of 15 days (±3 days). Convergent and diver-gent validity was assessed with Spearman rank correlation coeffi-cients of the SF-12 and HADS scores. Discriminative validity wasassessed by comparing SEALS scores among patient groups accord-ing to demographic and clinical characteristics and SF-12 mea-sures. Parametric and nonparametric tests were appliedaccording to variable characteristics.All  P   values were based on two-tailed tests, and statistical sig-nificance was set at  P   < 0.05. 3. Results  3.1. Patient demographic and clinical characteristics A total of 600 patients were recruited, of whom 595 were evalu-able for the analysis. Two patients were excluded because of theirage (<18) and three patients because they did not complete theSEALS questionnaires. Demographic and clinical characteristicsare summarized in Table 1.With respect to concomitant diseases and treatments in thepopulation analyzed, the most common were cardiovascular dis-eases ( n  = 64, 10.8%) and cardiovascular treatments ( n  = 65, 10.9%).According to the neurologists’ criteria, 9.9% of patients ( n  = 59)had presented with some kind of AED-related adverse event inthe preceding 4 weeks. Most patients (99.2%,  n  = 589) reportedgood compliance ( P 75%) with the AED treatment during the pre-ceding week.  3.2. SF-12 and HADS scores SF-12 Physical Component and Mental Component scores were48.9 (SD = 8.9, median = 50.8, range 19.9–62.5)and 47.3 (SD = 11.3,median = 50.6, range 13.9–65.5), respectively.For the HADS, the mean anxiety score was 6.8 (SD = 4.3) andmean depression score was 4.4 (SD = 4.2). Patients were classifiedas presenting anxiety or depression symptoms according to the fol-lowing cutoff scores: <8 = absence of symptoms, 8–10 = borderline Original version Spanish translation 1 Spanish translation 2 Items revision Spanish first consensus version English back–translation Comparison between srcinal and back–translated versions Spanish second consensus version Cognitive debriefing with 3 epilepsy patients Final pre-test version Fig. 1.  Translation and cultural adaptation process. V. Villanueva et al./Epilepsy & Behavior 14 (2009) 96–101  97  or mild disorder, 11–14 = moderate disorder, and >14 = presence of severe symptoms. The percentage of patients in each category foranxiety symptoms was as follows: absent 62% ( n  = 364), mild19.9% ( n  = 117), moderate 12.1% ( n  = 71), and severe 6% ( n  = 35).The percentage of patients in each category for depression scoreswas as follows: absent 78.6% ( n  = 458), mild 11% ( n  = 64), moderate7.2% ( n  = 42), and severe 3.3% ( n  = 19).  3.3. Descriptive analysis, feasibility, and internal structure of the SEALS  Scores were adjusted to a scale ranging from 0 (worst quality of life) to 100 (best quality of life). The mean score obtained for theSEALS Inventory was 60.7 (SD = 17.2, range 13.2–96.5) (Fig. 2).Minimal scores (floor effect) with respect to factors were: Worry,34 patients (5.9%); Temper, 14 patients (2.4%); and Tiredness, 5 pa-tients (0.9%). No maximum score (ceiling effect) was found for anyof the scale factors.There were statistically significant differences in SEALS with re-spect to demographic characteristics. Women scored lower thanmen ( P   = 0.003). Older patients (>45 years) scored lower thanyounger patients (35–44 years) ( P   = 0.0002). Patients with no for-mal education scored lower than patients who had attended uni-versity ( P   = 0.002). Regarding employment status, housewivesand retired patients scored lower than active patients and students( P   = 0.0009).With respect to feasibility, the most frequently unanswereditems were: item 32 (Have you felt slowed up? 3.5% of patients, n  = 21), item 35 (Have you felt slowed up or dulled in the companyof other people? 1.7% of patients,  n  = 10), and item 9 (Have youthought a lot about problems you may have? 1.5% of patients, n  = 9). Almost all patients (85.2%,  n  = 507) responded to all theitems on the questionnaire items, 10.3% of patients ( n  = 61) didnot complete one item, 2.4% of patients ( n  = 14) did not completetwo items, and 2% of patients ( n  = 13) did not complete three ormore (to a maximum of 13) items.Exploratory factor analysis led to the identification of five fac-tors: Cognition, Dysphoria, Tiredness, Temper, and Worry. Thequestionnaire structure was similar to that proposed by Gillhamet al. [12] in their srcinal proposal (Table 2); almost all items were present in the questionnaire structure (Table 2). With the excep-tion of items 22 (Have you found it difficult to enjoy yourself?)and 33 (Have you avoided mixing with people?), srcinally in-cluded in the Cognition factor and included in the Dysphoria factor  Table 1 Demographic and clinical characteristics Sex,% men 50.0Age, mean (SD) [range] 41.7 (14.8) [18–90]Educational level (%)No formal education 2.9Primary education 41.9Secondary school 35.9University 19.3Employment status (%)Actively working 55.6Studying 7.8Unemployed 8.3Housewife 15.7Retired 12.7Marital status (%)Married/living with partner 52.9Single 42.2Separated/divorced 3.2Widow 1.7Concomitant disease (%) 36.0Concomitant medication (%) 28.1Years since epilepsy diagnosis, mean (SD),median [range]16.7 (13.1), 14 [1–70]Seizures in the last 12 months, mean (SD),median [range]18.8 (169.4), 1 [0–4000]Type of seizures (%)Partial 73.4Generalized 25.6Unclassified 1.0Antiepileptic treatment (%)Monotherapy 55.6Polytherapy 44.4Treatment compliance in the last week (%) P 75% 99.2<75% 0.8 Fig. 2.  SEALS global and individual scale scores. The number of patients providing responses to all items included in each individual scale/overall scale score is indicated inparentheses.98  V. Villanueva et al./Epilepsy & Behavior 14 (2009) 96–101  in our analysis, and items 4 (Have you felt too tired to do anythingat all in the evening?), 6 (Have you felt alert even when on yourown?), 27 (Have you gone to bed earlier than you usually do?),and 36 (Have other people had to make decisions for you?), whichpresented correlations lower than 0.4 and were not related to anyfactor.  3.4. Reliability and reproducibility of the SEALS  Good internal consistency was indicated by a Cronbach  a  coef-ficient of 0.9330. For individual items, the coefficient ranged from0.9279 (Cognition) to 0.672 (Worry). Results are summarized inTable 3. Consistency in the responses was calculated by test–retestanalysis in a subgroup of 77 patients with stable medication. Theintraclass correlation coefficient (ICC) was calculated. The totalSEALS score showed excellent reliability, with a mean ICC above0.75 (Table 3).  3.5. Convergent and divergent validity of the SEALS  Analysis of the correlations between scores on the SEALS andHADS revealed negative correlations between the SEALS scoreand Anxiety and Depression scores, –0.66 and –0.71, respectively,indicating a worse perception of QoL.The five factor scores and the total score on the SEALS were pos-itively correlated with SF-12 summary scores and scores on theeight indices of the SF-12, indicating that the patient’s perceptionof a better QoL was associated with his or her perception of betterpsychosocial functioning (Table 4).  3.6. Discriminative validity of the SEALS  With respect to epilepsy-related factors, a poorer SEALS overallscore was associated with partial seizures, lack of seizure control,polytherapy, and AED-related adverse events. The presence andseverity of emotional disturbances, either anxiety or depressionsymptoms, were associated with a lower SEALS score, indicatingpoor perception of QoL. Lower SF-12 Physical and Mental summaryscores were also associated with lower total SEALS scores (Table 5).As the SEALS score was related to demographic characteristics,sex, age, educational level, and employment status were entered ascovariates in a univariate analysis. All statistically significant dif-ferences among groups with respect to epilepsy-related factors,emotional disturbances, and generic QoL remained.  Table 2 SEALS internal structure: factor analysis Spanish version UK version (Gillham et al. [12])Factor Item a,b Factor ItemCognition  5, 13, 14, 16, 18, 21, 23, 24, 25, 26, 28, 32, 35, 36, 38  Cognition 5, 13, 14, 16, 18, 21, 22, 23, 24, 25, 26, 28, 32, 33, 35, 36, 38Dysphoria  1, 8, 10, 11, 19, 22, 30, 33  Dysphoria 1, 6, 8, 10, 11, 19, 30, 37Tiredness  2, 7, 20  Tiredness 2, 4, 7, 20, 27Temper  3, 12, 29, 34  Temper 3, 12, 29, 34Worry  9, 15, 17, 31  Worry 9, 15, 17, 31 a Correlation factor–item >0.4 (ítems 4, 6, 27, and 37 showed a correlation <0.4) b In boldface are items corresponding to those in the srcinal version by Gillham et al. [12].  Table 3 Reliability: Internal consistency (Cronbach’s  a  coefficient) and reproducibility (intraclass correlation coefficient) n Cronbach’s  a  coefficient Intraclass correlation coefficient 95% CI a P b Min MaxAll items 507 0.9330 0.9208 0.8691 0.9521 <0.0001Cognition 536 0.9279 0.9169 0.8658 0.9485 <0.0001Dysphoria 571 0.7283 0.9091 0.8557 0.9428 <0.0001Tiredness 585 0.6808 0.8566 0.7738 0.9091 <0.0001Temper 585 0.7921 0.9270 0.8841 0.9540 <0.0001Worry 581 0.6720 0.7315 0.5764 0.8299 0.0003 a Ninety-five percent confidence interval. b Fisher’s F test.  Table 4 Correlation of SEALS Inventory with HADS and SF-12 Health Survey SEALS Inventory a Total score Cognition Dysphoria Tiredness Temper WorryHADSAnxiety –0.67 –0.58 –0.48 –0.42 –0.50 –0.49Depression –0.71 –0.66 –0.66 –0.45 –0.39 –0.32SF-12 summary scoresPhysical component 0.46 0.44 0.30 0.36 0.21 0.27Mental component 0.67 0.59 0.59 0.36 0.50 0.37SF-12 indicesGeneral Health 0.54 0.48 0.44 0.34 0.31 0.34Physical Functioning 0.39 0.38 0.28 0.30 0.15 0.20Role Physical 0.53 0.51 0.34 0.40 0.30 0.31Role Emotional 0.53 0.49 0.40 0.30 0.39 0.32Bodily Pain 0.37 0.35 0.23 0.27 0.22 0.26Mental Health 0.67 0.58 0.58 0.38 0.54 0.43Vitality 0.51 0.45 0.51 0.33 0.29 0.21Social Functioning 0.61 0.57 0.49 0.39 0.35 0.32 a All with correlation Spearman’s test P < 0.0001 (except Physical Functioning vs Temper, P = 0.0003). V. Villanueva et al./Epilepsy & Behavior 14 (2009) 96–101  99  4. Discussion The psychometric properties of the Spanish version of the 38-item SEALS Inventory tested in the present study were satisfactory.Our results reproduce the previous results reported for the English[13] and French [15] versions. Although the scores for the Spanish validation were slightly higher than those for the French version,both the overall score and the individual scale scores followedthe same pattern as those obtained in the French study, with ahigher score on the Dysphoria factor and a lower score on the Wor-ry factor.The overall score and the five factor scores on the SEALS Inven-tory showed adequate variability, ranging from 0 to 100 in approx-imately all scales. Our results indicated that neither overall norsingle scores reached the maximum (ceiling effect). Minimal scores(floor effect) were observed for Cognition, Tiredness, Temper, andWorry by less than 6% of patients.With respect to demographic characteristics, the SEALS scorewas associated with age, sex, educational level, and employmentstatus. A relationship with educational level and employment sta-tus had already been reported, not only for the SEALS question-naire [14,15], but also for many other health-related QoL  questionnaires [18–20]. On the contrary, the association with age and sex, not observed in previous studies [12,15], may be ex- plained by the higher percentage of emotional disturbances amongwomen and older patients and the clear relationship between theHADS and SEALS scores.The questionnaire was well accepted, and more than 80% of pa-tients completed all the items. As discussed by Gillham et al. [13],this compliance rate may be artificially increased, as the question-naires were completed in the clinic, where the neurologists requesttheir return when completed. However, when the compliance ratewas analyzed for those questionnaires completed at home by pa-tients who participated in the retest study, the percentage of pa-tients who completed all the items was also greater than 80%( n  = 63 patients).The internal structure of the questionnaire almost reproducesthe five-factor structure proposed by Gillham et al. [12].To assess the internal consistency of the SEALS Inventory, Cron-bach’s  a  coefficient was computed for each item as well as for thefive individual scales and the overall scale. Because it has been sta-ted that values greater than 0.7 are required to consider internalconsistency [21], our results indicate that the Spanish version of the SEALS has good internal consistency. Again, the same correla-tion pattern was observed between ours and the French validation,which reported a lower Cronbach a coefficient for the Worry (0.65)and Tiredness (0.7) factors. Lower coefficients for the Worry factorwere also reported by Gillham et al. [13].The data obtained from the subgroup of 77 patients who com-pleted the questionnaire twice revealed that the Spanish versionof SEALS has good reproducibility. The results reported in theFrench study were lower compared with ours; as the authors men-tioned, this may be due to the inclusion of patients whose medica-tion had been changed, as AED treatment remained stable betweenthe first and second questionnaires for all of our patients.The SEALS global and individual scale scores showed significantcorrelation with both the HADS and SF-12. Although in the FrenchSEALS validation study, the authors [15] administered QoL ques-tionnaires different from ours, the patterns of correlations ob-served in the studies were similar: lower correlations amongSEALS scores and physical indices and higher correlations withemotional, social, and cognitive indices. These results indicate thatthe SEALS Inventory has good convergent and divergent validity.The SEALS Inventory showed good discriminative validity, dem-onstrating the ability to discriminate patients according to differ-ent epilepsy-related factors and to differentiate clearly amongpatients according to the presence and severity of emotional dis-turbances. The influence of these factors on the SEALS score wasindependent of demographic characteristics, as the covariate anal-ysis revealed.Because there are several Spanish-speaking populations locatedin different countries around the world, caution on the use of thepresent Spanish version is warranted. Cross-cultural adaptationof any self-report measure entails not only the linguistic adapta-tion, that is, translation of the instrument, but also cultural adapta-tion. This cultural adaptation results in an instrument that isculturally relevant and significant to a specific population andshould be performed when the questionnaire is going to be usedin a different country with the same language [22]. Therefore,the Spanish version of the SEALS Inventory could be used in otherSpanish-speaking countries after cultural adaptation.In conclusion, validation of the Spanish version of the SEALSInventory revealed adequate psychometric characteristics, withgood levels of feasibility, reliability, and discriminative validity,similar to those obtained previously with the UK and French ver-sions. Therefore,this validatedscale can be used in epilepsystudiesand clinical trials with antiepileptic drugs to capture aspects re-lated to adverse effects of antiepileptic treatment on cognitionand behavior.  Acknowledgments This study was supported by an unrestricted grant from EisaiFarmacéutica S.A. The authors thank Dr. Ruth Gillham for her col-laboration in adaptation of the questionnaire.  Table 5 Discriminative validity with respect to epilepsy-related factors, emotional distur-bances, and generic QoL  n a Mean SD P b Type of epileptic seizure c Partial 364 58.7 17.4 0.0003Primary generalized 137 65.6 15.9Unclassified seizure 5 66.3 17.5Number of seizures in preceding week d None 412 61.8 17.0 0.0098One 42 56.3 17.1Two or more 41 54.9 16.6Antiepileptic drug treatmentMonotherapy 285 64.3 16.8 <0.0001Polytherapy 222 56.0 16.8AED-related adverse effectsAbsence 454 61.6 17.1 0.0004Presence 53 52.9 16.6Severity of anxiety (HADS) e Normal 305 67.9 14.1 <0.0001Mild 102 56.7 13.8Moderate 61 45.2 12.3Severe 34 35.0 13.6Severity of depression (HADS) e Normal 383 66.1 14.5 0.0024Mild 60 46.4 13.9Moderate 36 40.5 12.1Severe 18 34.4 11.6Physical Component summary (SF-12)<50 219 53.2 16.1 <0.0001>50 261 67.5 15.3Mental Component summary (SF-12)<50 240 51.4 15.2 <0.0001 P 50 240 70.5 13.3 a Analyses were performed in patients with available data. b Mann–Whitney or Kruskal–Wallis test. c SEALS score: partial < primary generalized. d SEALS score: none > one; none > two or more. e SEALS score: normal > mild > moderate > severe.100  V. Villanueva et al./Epilepsy & Behavior 14 (2009) 96–101
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!