Validity and Reliability Test on Sleep Quality Scale (SQS) Instruments in Indonesia Version on Cancer Patients

Amzal Mortin Andas1,2, Christantie Effendi*3, Sri Setyarini4 1Department Medical Surgical Nursing, Sekolah Tinggi Ilmu Kesehatan Bani Saleh, Bekasi, Jawa Barat 17113, Indonesia 2Department of Nursing, Lincoln university college, Lembah Sireh-15050, Malaysia 3Department Medical Surgical Nursing, Gadjah Mada University, Yogyakarta 55281, Indonesia 4Department of basic and Emergency nursing, faculty of medicine, Public Health and Nursing, Gadjah Mada University, Yogyakarta 55281, Indonesia


INTRODUCTION
Cancer is one of the leading causes of death worldwide. According to GLOBOCAN data, the International Agency for Research on Cancer (IARC) found that in 2012 there were 14,067,894 new cases of cancer and 8,201,575 deaths from cancer worldwide (Siegel et al., 2015). In Indonesia, cancer is the cause of death of about 8.2 million people (Kemenkes, 2015). The prevalence of cancer patients in the population of all ages in Indonesia, with a majority of 1.4%. Cancer patients generally experience pain, fatigue, distress, and sleep disorders (Trill, 2013). Based on literature studies, over 60% of cancer patients have problems with their sleep disorders (Mercadante et al., 2004).
Every year, estimated that there are 20% -70% experiences sleep problems, and about 17% are diagnosed with severe sleep disorders (Li et al., 2017). Low sleep quality affects the increased risk of cancer until death (St-Onge et al., 2016), declining physical health, psychology, and quality of life (Harvey et al., 2008).
Sleep quality measurement can be carried out by various methods, such as subjective and objective methods. One of the objective methods is polysomnography; it is the standard gold of sleep quality measurement (Schabus et al., 2014).
The most common instrument used for measuring sleep quality in clinical patients is the PSQI instrument for measuring sleep quality of breast cancer patients (Effendy et al., 2015). Another instrument that can be used, such as the Sleep quality scale (SQS) developed by Yi et al. (2009). This instrument uses for a comprehensive assessment of sleep quality in the general population. Nevertheless, SQS instrument has never been tested both of the validity and reliability in clinical patients, especially cancer patients in Indonesia, so that it is essential to research with aiming to test the validity and reliability of SQS instruments in cancer patients in which the result of this research can be utilized on sleep quality and more comprehensive on sleep quality assessment of cancer patients.

Research Method
The SQS instrument consists of six components: daytime dysfunction, restoration after sleep, insomnia, dif iculty waking, sleep satisfaction, and dif iculty maintaining sleep. Also, the SQS instrument consists of 28 items covering questions about symptoms of insomnia. SQS instruments using a Likert scale with ive choices of answer those are Not at all (0), Little (1), Good Enough (2), Good (3), and Very Good (4). Respondents will mark the checklist on one the answer from 0 to 4 responses according to the respondent's condition. The score range is 0 to 84, with a higher score indicating a lower quality of sleep. Research about the validity and reliability test on this instrument has not been held in Indonesia. Therefore, before doing the validity and reliability test, the researcher conducted the translation process into SQS Indonesian version. The method of translation is undertaken by using Brislin's theory. After translating into Bahasa Indonesia, the researcher then asks for an opinion and assesses the translated SQS instrument. Furthermore, researchers will conduct back translation with different translators.

Sample
The populations in this study were male and female patients under the age of 18 -65 who were diagnosed with cancer by doctors at RSUD Prof. Dr Mar-gonoSoekardjoPurwokerto. Sampling was determined by non-probability sampling in consecutive sampling by taking all patients who meet the criteria until the numbers of samples are met.
The numbers of a sample are based on the literature to test the validity and reliability of health problems. The sample required based on Rules of Thumb determination that the number of instrument items multiplied by 3. Therefore, The sample size in this study was 90 people.

Reliability Validity
The content validity test carried out SQS validity instrument test was measured its coef icient relevance using Aiken's V Construction validity measured with Pearson product-moment, with signi icance level 0,05 and R table with several respondents 90 that was 0.205. Internal consistency reliability test using Cronbach Alpha with coef icient α> 0.7 indicated that the item was reliable.
The reliability test of retest using correlation coeficient test between the total score of respondents' answers during the initial examination and the second test. Researchers also conducted a correlation test to measure the relationship between factors.

RESULTS
In this study, the majority of respondents aged were between 46 -55 years old. All of the respondents consisted of males and females. Of the respondents, 33.33% were male, and 52.2% of respondents were female. Most respondents were selfemployed (36.7%), and 64.7% of respondents had income below Rp. 1.400.000.
The disease's condition was dominated by nasopharyngeal cancer (36.7%), with illness duration, the majority of fewer than 12 months (86.7%). Most research respondents had undergone chemotherapy were 82.2%, and only one respondent had ever taken sleeping pills. Demographic characteristics could be seen in Table 1.

Validity
In this study, the assessment of the content validity coef icient was based on Aiken's V formula. The Internal consistency testing using Cronbach's Alpha and reliability test of retest using correlation coef icient between 2 answers from the same respondent but at a different time. In this study, the time interval between the irst test and the second test was two weeks. The number of respondents to do a retest was 31 respondents. Internal consistency reliability results could be seen in Table 3.
In the internal consistency reliability test, the value of Cronbach alpha was 0.849. This value was considered consistent in the instrument to be used was more than 0.70, and it was considered very good if the cost was more than 0.90. The results of the retest on the SQS instrument SQS Indonesian version in 31 respondents showed an excellent stability level; it was 0.903. The instrument was considered had good reliability if the ICC value was 0.75 to 0.9.

DISCUSSION
The calculation result of the content validity SQS instrument revealed that the SQS instrument had an excellent relevance to the concept of sleep quality. The overall item of the SQS instrument got a value of Aiken equal to 0,976. All of the tools SQS belongs to an excellent category or has good content relevance to measuring sleep quality because Aiken's value V was closed to 1. It was indicated from the calculation of each item in the SQS instrument.
The number of SQS Instrument was 28 items and only four items whose value was below 1. However, those four items were still maintained because the cost was high enough that was 0.833.
The results of calculating the original SQS instrument's content validity showed that the values obtained were not much different because the original instrument used the CVI formula to assess the content validity.
The result of r value in all items 1-28 was in the range of 0.220-0.561. Based on this result, it could be considered that the SQS instrument was valid since all of the respondent's answers have high diversity so that r value was signi icant.
Furthermore, the reliability test results for all items of the SQS instrument were stated very reliable with Cronbach's Alpha value 0.849. This result has the same consistency level as the original instrument of SQS that was 0,87, while the retest value was 0,903. This was indicated by the SQS instrument consistency level and good stability. This result showed a good similarity of consistency with the original SQS instrument of 0.87 (Yi et al., 2009).
This research also performed a reliability test of retest with an interval time between the irst and second test was 1-2 weeks considering to get the From the results of the retest, the value obtained 0.903. It showed that the score of the SQS instrument in retest has high stability. This could be due to the distance of time being relatively close, so respondents could still remember the irst test answers.

CONCLUSION
Regarding the discussion above, the researcher concluded that the Sleep Quality Scale instrument that had been translated into Indonesian and tested either the validity or reliability in cancer patients at RSUD Prof.Dr. Margono Soekaedjoindicated to be valid and reliable. SQS instruments had an adequate level of internal consistency and stability.

ACKNOWLEDGEMENT
The researcher thanked all the research respondents involved in this study and the entire Research Team (Data collector, Massage technician, and Hospital Nurse).

Con lict of Interest
The authors declare that they have no con lict of interest for this study.

Funding Support
The author declared that they have no funding support for this study.