questionnaire (FOSQ) to Turkish population

(1)

questionnaire (FOSQ) to Turkish population

Bilgay İZCİ¹, Hikmet FIRAT¹, Sadık ARDIÇ¹, Oğuz KÖKTÜRK², Ethem GELİR³, Meltem ALTINÖRS¹

1SSK Ankara Eğitim Hastanesi Göğüs Hastalıkları, Uyku Merkezi,

2Gazi Üniversitesi Tıp Fakültesi, Göğüs Hastalıklara Anabilim Dalı, Uyku Merkezi, Ankara 3Karaelmas Üniversitesi Tıp Fakültesi, Fizyoloji Anabilim Dalı, Zonguldak

ÖZET

Uykunun fonksiyonel sonuçları ölçeğinin (FOSQ) Türk toplumuna uyarlanması

Uyku anketinin fonksiyonel sonuçlarının Türk versiyonu (FOSQ.tr), obstrüktif uyku apne hipopne sendrom (OUAHS)’lu hastalarda psikometrik özellikleri yönünden değerlendirildi. FOSQ testi önce Türkçe’ye çevrilip, sonra tekrar İngilizce’ye çevrildi. Psikometrik değerlendirme amacıyla, ardarda seçilmiş 73 hasta ile birlikte 73 kontrol olgusuna test uygulandı.

Testlerin geçerlilik ve güvenilirlik çalışmaları yapıldı. FOSQ.tr’nin tümünde (α= 0.92) ve alt skalalarında (α= 0.76-0.80) Cronbach’s alfa değerleri, anketin iç güvenilirliğinin tam olduğunu düşündürdü. FOSQ.tr’nin test, retest geçerliliği sosyal sonuçlardan farklı olarak, tüm skorlar (r= 0.7) ve alt skalalar için anlamlıydı (r= 0.5’den 0.8’e kadar, toplamda p< 0.01).

FOSQ.tr Epworth uykululuk skalası ile korelasyon gösteriyordu; korelasyon katsayısı alt skalalar için r= -0.5’den-0.62’ye kadar (p< 0.05) ve tüm skor için r= 0.64 (p< 0.01) olarak değişiyordu. Diskriminant analiz sonucu FOSQ.tr’nin normal bi- rey ile hastanın anlamlı derecede ayırt edilebileceğini göstermiştir (p< 0.03). Sonuç olarak, FOSQ.tr’nin psikometrik özel- likleri davranışlarda gündüz aşırı uyku halinin etkilerinin saptanmasında geçerli ve güvenilir bir test olduğu gösterilmiştir.

Anahtar Kelimeler: Gündüz aşırı uyku hali, yaşam kalitesi, anket, güvenilirlik, geçerlilik.

SUMMARY

Adaptation of functional outcomes of sleep questionnaire (FOSQ) to Turkish population

Izci B, Firat H, Ardic S, Kokturk O, Gelir E, Altinors M

SSK Ankara Educational Hospital, Department of Chest Diseases and Sleep Center, Ankara, Turkey.

Turkish version of functional outcomes of sleep questionnaire (FOSQ.tr) was examined for its psychometric properties in patients with obstructive sleep apnoea hypopnoea syndrome (OSAHS). The FOSQ was translated into Turkish using a for- ward-backward translation. For the psychometric evaluation, 73 consecutive patients were selected along with 73 control subjects. Internal consistency, test-retest reliability, concurrent and discriminant validity were investigated. Values of Cron- bach’s alpha for the total FOSQ.tr (α= 0.92) and its sub-scales (α= 0.76-0.80) suggest that the questionnaire was consistent internally. Test-retest reliability of the FOSQ.tr was significant for the total score (r= 0.7) and the sub-scales apart from so-

Yazışma Adresi (Address for Correspondence):

Dr. Hikmet FIRAT, SSK Ankara Eğitim Hastanesi Göğüs Hastalıkları Uyku Merkezi, ANKARA - TURKEY e-mail: hikfirat@ttnet.net.tr

(2)

Excessive daytime sleepiness (EDS) deteriora- tes cognitive and psychological well being and is a potential risk factor for traffic and work accidents (1-5). Thus it may have a serious impact on functional outcomes relevant to daily behaviour and quality of life. EDS most often results from obstructive sleep apnoea hypopnoea syndrome (OSAHS) (6-10). It occurs in a vari- ety of sleep disorders such as narcolepsy, idi- opathic hypersomnia and periodic limb move- ment disorder (PLMD), etc, but impairment in many of the areas of daytime functions improves with treatment of these illnesses (6,8,11- 13). However it is difficult to assess the degree of the impairment of functional outcomes in patients with sleep disorders due to the fact that it is such a broad and subjective area. Although a large number of self-reported scales have been produced to quantify the degree of sleepiness, these scales only focus on the sleepiness state.

Functional outcomes of sleep questionnaire (FOSQ) is designed to capture the impact of EDS on performance of multiple activities of daily living and the extent to which these abilities are improved by effective treatment of EDS (14,15).

Standardised and well-tested instruments such as FOSQ also are needed in order to evaluate the effect of EDS on the performance of daily activities in the Turkish population. Therefore, the aim of this study was to produce a Turkish version of the FOSQ (FOSQ.tr) to examine its psychometric properties in Turkish population.

MATERIALS and METHODS Study Participants

Individuals consecutively attending the Sleep Centre of SSK Training Hospital and the Depart- ment of Sleep Disorders of Gazi University Medi- cal Faculty for diagnostic polysomnography we-

re invited to participate to the study and all agre- ed. Seventy-three of them who had apnoea hypopnoea index (AHI) ≥ 5 were selected as a patient group. Forty-six of them who had AHI <

5 and no EDS [Epworth sleepiness scale (ESS):

7 sd ± 5], and an additional 27 subjects, mainly hospital employees, who worked during the daytime and had no experience of EDS, OSAHS or narcolepsy were chosen as a control group. The exclusion criteria for both groups were having a medical history of blindness or a major mental illness (e.g. depression). The inclusion criteria for both groups were completing the FOSQ.tr.

without any assistance.

In addition, for test-retest reliability, 58 third-ye- ar medical students were recruited. All were he- althy at the study time. The protocol was appro- ved by the hospitals’ ethics committee.

Procedure and Measurements

After providing informed consent, patients and control groups were asked to complete FOSQ.tr, ESS and a brief demographic questionnaire inc- luding questions on age, martial situation and education. Patients completed the questionnaires in the sleep laboratory either prior to or upon the completion of their routine nocturnal diagnostic sleep study. All had height and weight recorded.

FOSQ.tr was applied to 58 medical students participating in a course. Thirty of them, who were available, were re-tested 4 weeks later.

Sleepiness-ESS

The ESS is a simple, eight-item self-administered scale which is widely used in clinical practice to quantify the level of daytime sleepiness in situations of different soporificity. It has a total score range of 0-24 and scores > 10 are indica- tive of a high level of daytime sleepiness (16).

cial outcome (r= 0.5 to 0.8, all p< 0.01). FOSQ.tr correlated moderately with Epworth sleepiness scale (ESS), with coefficients ranging from r= -0.5 to -0.62, (all p< 0.05) for the sub-scales, and r= -0.64 (p< 0.01) for total score. Discriminant analysis showed that FOSQ.tr could significantly distinguish the patients from normal subjects (p< 0.03). The psychometric properties of the FOSQ.tr suggest that it is a valid and reliable instrument for the assessment of the impact of disorders of excessive sleepiness on daily behaviour.

Key Words: Excessive daytime sleepiness, quality of life, questionnaires, reliability, validity.

(3)

FOSQ

FOSQ is designed to measure the impact of EDS caused by sleep-disorders on physical, mental and social functioning in everyday activities, and can also be used to determine how treatment improves the quality of a person’s life (14,15). It consists of 30 questions which repre- sent five sub-scales: activity level, vigilance, sexual activities, general productivity and social outcome (14).

FOSQ was applied in a decreased 26-item version (dropping the sexual activities sub-scale) in the current study. Although the sexual activities sub-scale is a special characteristic of the FOSQ, it was found in our pilot study that this scale was not acceptable for the particular middle-aged Turkish population.

Internal consistency of the original FOSQ ranged from 0.86 to 0.91 for the subscales and was 0.95 for the total score. The test-retest reliability ranged from r= 0.81 to r= 0.90 for the five subscales and r= 0.90 for the total score (14). These psychometric outcomes demonstrated that the original FOSQ’s parameters were acceptable for its application in research and in clinical practice to measure functional status outcomes for persons with EDS (14).

Scoring

The questions have graded responses on a 4- point scale [no difficulty, a little, moderate, or extreme]. There is a mean-weighted item score for each sub-scale with a range of 0-4; sub-scale scores are totalled to produce a global score.

In this study, the total score ranged from 4 to 16 due to excluding the sexual activities sub-scale.

Lower scores are associated with greater dysfunctionality (14).

FOSQ’s Translation to Turkish (Content Validity) Firstly, FOSQ was translated into Turkish by a specialist in social psychology and two other clinicians in sleep disorders area and the final text was constructed using their common sentences.

Secondly, the text was administered to a small group of seven patients with OSAHS. After this

administration, it was decided that the intimacy and sexual relationships sub-scale, which inclu- des 4 questions (difficulty in sexual arousal, sex desire, orgasm and intimacy due to sleepiness), should be excluded, since they were not answe- red (14). In addition to this, some words which caused misunderstanding were changed. Next, another translation from Turkish back to English was carried out by a bilingual professional trans- lator to compare with the original text. Finally, some changes were made so that it would be understandable and easy to fill out.

Statistical Analysis

Data analyses were performed using SPSS 10^th version for Windows (SPSS inc., Chicago, IL, USA). The t-test was performed to evaluate differences in demographic and clinical characteristics of the study sample, and group differences for discriminant analysis. The analysis of va- riance was performed, followed when appropriate by the Tukey multiple comparison test to discriminate patients with different severities of illness. Chi-square technique was used for cate- gorical variables. Internal reliability was assessed using Cronbach’s alpha. The Pearson correlation coefficient was calculated to assess the relationship between ESS and FOSQ.tr, and test-retest reliability. p-values less than 0.05 were considered as statistical significant.

RESULTS Study Populations

Demographic and clinical characteristics of the study sample are presented in Table 1. Patients were not different from the control group in terms of age, education level and marital situation but they were predominantly male (p= 0.001) and had a higher body mass index (BMI) than controls (p< 0.02) (Table 1). However BMI are not correlated with total FOSQ.tr and its sub-scales scores in both patients and control groups (p> 0.05).

Patients had also significantly higher AHI (p< 0.001) and ESS total score than the controls (p< 0.03) (Table 1). The mean age of the group of medical students recruited for test-retest reliability was 22sd2. Seventy-three percent of them were male.

(4)

Reliability of FOSQ.tr

Internal consistency: The internal consistency of variables was examined using Cronbach’s alpha. Cronbach’s coefficients were calculated for the total 26-item questionnaire and for its sub-scales. Cronbach’s coefficients ranged from 0.76 to 0.80 across four sub-scales and 0.92 was found for total score (Table 2).

The test-retest reliability: The FOSQ.tr and its sub-scales’ scores of 30 subjects over a period of 4 weeks were compared to assess the test-retest reliability of the questionnaire. There were moderate correlations between two measurements apart from social outcome sub-scale (total score: r= 0.7, p< 0.01; activity level: r= 0.73, p< 0.01; vigilance: r= 0.8, p< 0.01; general productivity: r= 0.5, p< 0.01; social outcome: r=

0.3, p> 0.05).

Validity of FOSQ.tr

Concurrent validity: The Turkish version of ESS was used to assess the concurrent validity of FOSQ.tr. There were moderate significant relationships between ESS and the total FOSQ and its sub-scales, with coefficients ranging from r=

-0.5 (social function) to r= -0.53 (general productivity) to r= -0.62 (both activity level and vigilance), (all p< 0.05), and r= -0.64 (p< 0.01) for total score.

Discriminant validity: Discriminant validity analyses of FOSQ.tr were performed on the FOSQ.tr’s sub-scale-scores and total-scores of the patients, controls groups and separately on patients divided into 3 groups according to their severity of illness (mild: AHI 5-14/hr, moderate: AHI 15-30/hr and severe: > 30/hr OSAHS).

A comparison between patients’ and controls’

FOSQ.tr scores showed that the total FOSQ.tr Table 1. Subjects’ demographic features.

Patients (n= 73) Controls (n= 73) p

Mean (SD) age, (years) 48 ± 12 45 ± 11 > 0.2

Sex (%) 16% (12) female %41 (30) female 0.001

%84 (61) male %59 (43) male

Marriage (%) %93 married %81 married > 0.1

Education (%) %72 ≥ high school %65 ≥ high school > 0.8

%28 < high school %35 < high school

Mean (SD) BMI (kg/m²) 31 ± 5 27 ± 4 < 0.02

Mean (SD) AHI, events/hr 37 (30) 2 (1) < 0.001

Mean (SD) ESS 14 (6) 8 (5) < 0.02

SD: Standard deviation, BMI: Body mass index, AHI: Apnoea-hypopnoea index, ESS: Epworth sleepiness scale.

Table 2. Descriptive statistics and Cronbach’s coefficients calculated for the whole questionnaire and its sub- groups (n= 146).

Potential Obtained Cronbach’s Range: Item to

Mean SD range range alpha total correlation

Activity level and productivity (9 items) 2.85 0.67 0-4 1.22-4 0.80 0.15-0.68

Vigilance (7 items) 2.36 0.92 0-4 0-4 0.80 0.43-0.65

Social interactions (8 items) 3.03 0.73 0-4 0.88-4 0.80 0.46-0.60

Social outcome (2 items) 2.91 0.95 0-4 0-4 0.76 0.62

Total 11.20 2.72 0-16 2.99-16 0.92 0.22-0.68

SD: Standard deviation.

(5)

(p< 0.001) and all sub-scales (all p< 0.03), could significantly distinguish the patients from controls (Table 3). However, there were no significant differences between sub-groups of patients according to their severity of illness (p> 3).

DISCUSSION

This study provides the psychometric evidence that FOSQ.tr is a sufficiently reliable and valid measure for evaluating impairment of functional outcomes and quality of life due to EDS in Tur- kish patients with OSAHS.

FOSQ.tr has convincing internal consistency which is close to levels reported for the original study and another adaptation study (14,17).

FOSQ.tr had satisfactory test-retest reliability, except for social outcome. The current study failed to find test-retest reliability for the social outcome sub-scale, perhaps because of subjects used, who were all medical students with busy schedules and may not have had sufficient time and suitable place for having guests over/visi- ting people. Thus the total score and all sub- scales scores may be reproducible from one test to the next in middle-age population with EDS but not with students.

Functional status outcome has been described as “a multifaceted concept that characterizes the ability to meet needs, fulfil roles and mainta- in health and well-being” (18). In this study to examine validation of the FOSQ, sleep-related functional status was assessed comparing FOSQ.tr score to ESS. ESS correlated moderately with total FOSQ.tr and its all sub-scales.

These results were anticipated because the two scales evaluate daytime sleepiness from diffe-

rent aspects; the ESS quantifies the likelihood of sleeping in a given situation, while the FOSQ measures the effect of sleepiness on functional status outcomes. This finding indicated that EDS worsens with respect to functional status and the quality of life in all areas measured in subjects with EDS. The most noticeable degree of impairment occurred in the activities level (housework, sports, activity in the evening, activity in the morning, activity in the afternoon, ke- eping pace with others, relationship with fa- mily/friends) and vigilance (enjoying theatre, concert, watching television, driving a car for long/short distances, participating a meeting).

This finding is compatible with other studies that associated cognitive and psychofunctional impairment with sleepiness (1,2,19). Results from this study also comply with the adaptation study in Spanish and a study performed in on the elderly using FOSQ. These studies confirmed that the areas most effected by sleepiness were vigilance and those activities mentioned in FOSQ (17,20).

In this study, patients with OSAHS had significantly worse scores for total and all sub-scales of the FOSQ.tr. This result indicates that the sub-scales and total FOSQ.tr can successfully discriminate patients from control subjects and evaluate daily behaviour and quality of life in Turkish patients. However there were no significant differences in total FOSQ.tr and the sub- scales between sub-groups of patients concer- ning their severity of illness. This shows FOSQ.tr can not distinguish in this area.

The main difference between the Turkish and the original English version of the FOSQ is that

Table 3. Discriminant validity analysis on the FOSQ.tr sub-scale-scores and total-scores of patients and control subjects [Mean (SD)].

Patients (n= 73) Controls (n= 73) p

Activity level 2.6 (0.6) 3 (0.7) < 0.001

Vigilance 2.1 (0.9) 2.5 (0.9) < 0.03

General productivity 2.8 (0.7) 3.2 (0.7) 0.001

Social outcome 2.6 (1) 3.2 (0.8) < 0.001

FOSQ total 10.2 (2.6) 12 (2.5) < 0.001

SD: Standard deviation.

(6)

the former did not include sexual activities sub- scale which had to be excluded because of the feedback from the respondents who joined but failed to complete this sub-scale in the pilot study; because it was not appropriate for them to answer such private questions. FOSQ.tr would be more valuable if it included the sexual activities sub-scale which measures these activities affected by OSAHS and generally overlo- oked. However, our determination is consistent with earlier observations by other researchers in the Department of Sleep Medicine in Edinburgh University that Scotland’s middle-aged population found its content too intrusive, and refused to participate in it as well (18). This was also consistent with a study of 184 subjects older than 65, only 30.4% of whom filled out that section in USA (21).

The study has several limitations. First, the excessive sleepiness of the patients could have been demonstrated objectively by Multiple Sleep Latency Test (MSLT) or Maintenance of Wakeful- ness Test (MWT) rather than using ESS. Thus, it is likely that some subjects may not be aware of their EDS, whereas others may have an exagge- rated sense of it. Even better evidence about the validity of the FOSQ.tr could have come from considering the effect of treatment on FOSQ.tr scores, but this deficiency will have to be filled by a later study.

Second, the patients had higher BMI compared to controls, although the FOSQ.tr scores of these patients were not correlated with BMI (p> 0.05). Also patients group were predominantly male. If the control and patient groups were matched for BMI and gender, it would have been preferable. Howe- ver, it was difficult to find a person with high BMI who did not experience symptoms of OSAHS.

In conclusion, the psychometric characteristics of FOSQ.tr showed that it is a reliable and valid instrument to assess the impact of excessive sleepiness on multiple activities of daily life in Turkish patients with OSAHS. Thus, it can be a useful tool in clinical practice and research. Ho- wever further studies measuring EDS objectively along with MSLT or MWT and studying the effect of treatments on FOSQ.tr’s scores would be worthwhile undertakings.

ACKNOWLEDGEMENTS

The authors would like to thank Dr. T. Weaver, who developed the FOSQ, for giving us permis- sion to use the questionnaire and for her assess- ments of the Turkish and English versions of the FOSQ which we used in our study. Thanks are also due to Dr. H.M. Engleman and Professor N.J. Douglas for their support and encourage- ment. We would also like to express our appre- ciation to those who gave their time and effort to participate in this study.

REFERENCES

1. Cheshire K, Engleman HM, Deary IJ, et al. Factors impa- iring daytime performance in patients with sleep apnea/hypopnea syndrome. Arch Intern Med 1992; 152:

538-41.

2. Kingshott RN, Engleman HM, Deary IJ, Douglas NJ. Do- es arousal frequency predict daytime function? Eur Res- pir J 1998; 12: 1264-70.

3. Young T, Blustein J, Finn L, Palta M. Sleep-disordered breathing and motor vehicle accidents in a population- based sample of employed adults. Sleep 1997; 20: 608- 13.

4. Masa JF, Rubio M, Findley LJ. Habitually sleepy drivers have a high frequency of automobile crashes associated with respiratory disorders during sleep. Am J Respir Crit Care Med 2000; 162: 1407-12.

5. Horne J, Reyner L. Vehicle accidents related to sleep: A review. Occup Environ Med 1999; 56: 289-94.

6. Jenkinson C, Stradling J, Petersen S. Comparison of three measures of quality of life outcome in the evaluation of continuous positive airways pressure therapy for sleep apnoea. J Sleep Res 1997; 6: 199-204.

7. Firat H, Ardic S, Gelir E, et al. The relationship between arousal index and Epwoth sleepiness scale among patients with suspected obstructive sleep apnea syndrome.

J Sleep Res 2000; 9(Suppl 1): 131.

8. Douglas NJ. Clinicians’ Guide to Sleep Medicine. Oxford University Press, 2002: 5-26, 123-4.

9. Gottlieb DJ, Whitney CW, Bonekat WH, et al. Relation of sleepiness to respiratory disturbance index: The sleep he- art health study. Am J Respir Crit Care Med 1999; 159:

502-7.

10. Stradling JR, Crosby JH, Payne CD. Self reported sno- ring and daytime sleepiness in men aged 35-65 years.

Thorax 1991; 46: 807-10.

11. Winkelman JW, Carskadon MA, Doghramji K, McCarley RW. Excessive Daytime Sleepiness: Evaluation and Ma- nagement in Psychiatry. [Online]. The American Psychi- atric Association (APA), 2004 Jan, Available from:

URL:http://www.psych.org/cme/apacme/high- lights/excessiveDaytimesleepiness.pdf.

(7)

12. Engleman HM, Martin SE, Deary IJ, Douglas NJ. Effect of continuous positive airway pressure treatment on daytime function in sleep apnoea/hypopnoea syndrome. Lan- cet 1994; 343: 572-5.

13. Kingshott RN, Vennelle M, Hoy CJ, et al. Predictors of improvements in daytime function outcomes with CPAP therapy. Am J Respir Crit Care Med 2000; 161: 866-71.

14. Weaver TE, Laizner AM, Evans LK, et al. An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep 1997; 20: 835-43.

15. Montserrat JM, Ferrer M, Hernandez L, et al. Effective- ness of CPAP treatment in daytime function in sleep apnea syndrome: A randomized controlled study with an optimized placebo. Am J Respir Crit Care Med 2001; 164:

608-13.

16. Johns MW. Reliability and factor analysis of the Epworth sleepiness scale. Sleep 1992; 15: 376-81.

17. Ferrer M, Vilagut G, Monasterio C, et al. Measurement of the perceived impact of sleep problems: The Spanish version of the functional outcomes sleep questionnaire and the Epworth sleepiness scale. Med Clin (Barc) 1999; 113:

250-5.

18. American Thoracic Society/American Sleep Disorders As- sociation. Statement on health outcomes research in sleep apnea. Am J Respir Crit Care Med 1998; 157: 335-41.

19. McFadyen TA, Espie CA, McArdle N, et al. Controlled, prospective trial of psychosocial function before and after continuous positive airway pressure therapy. Eur Respir J 2001; 18: 996-1002.

20. Gooneratne NS, Weaver TE, Cater JR, et al. Functional outcomes of excessive daytime sleepiness in older adults. J Am Geriatr Soc 2003; 51: 642-9.

21. Gooneratne NS, Weaver TE, Maislin G, et al. A comparison of the Epworth Sleepiness Scale and the Functional Outcomes of Sleepiness Questionnaire in the assessment of excessive daytime sleepiness in the elderly. Sleep 2001; 24(Suppl): A233.