INTRODUCTION
Many studies have reported a prevalence of insomnia symptoms ranging 36%-91% in patients with alcohol use disorder [1]. Insomnia causes repeated drinking and may trigger workplace and interpersonal conflicts, decreased quality of life, and suicidal ideation [2]. Hospitalized patients with alcohol use disorder have increased rapid eye movement (REM) sleep, but sleep disorders repeatedly appear after discharge. The biggest problem causing recurrence of this disease is insomnia symptoms, which lead to repeated drinking; however, there is a lack of interest and research on this topic [3]. Moreover, complaints of sleep problems by patients with alcohol use disorder are often ignored [4]. Recently, the most common treatment for alcohol use disorder patients in Korea has been drug treatment;sleep interventions are rarely implemented [5]. Mental health professionals may believe that sleep problems are included in psychosocial treatments aimed at achieving sobriety; however, they may be considered as being in a blind spot where they are not implemented.
Conversely, scientific and physiological studies on the relationship between alcohol consumption and sleep are active [1,6]. Many studies have reported a strong relationship between alcohol consumption and sleep, and patients with sleep disorders may develop alcohol-related disorders conversely [7].
Recently, in South Korea, 59 addiction management integrated support centers have been operated by region, along with nine alcohol-designated hospitals [8].Among over 400 psychiatric hospitals, nine alcohol hospitals are designated by the Ministry of Health and Welfare across the country, which is interpreted as an unusual situation considering the high prevalence rate [9]. This suggests that patients may spend significant time at home. However, insomnia, which frequently occurs after discharge, leads to incorrect coping, including drinking, which repeats the vicious cycle of addiction [10].
Alcohol consumption due to sleep disorders after discharge affects sleep-related hormones and the central nervous system, interfering with normal sleep [7]. Active use of substances increases sleep maintenance insomnia [11,12], but active care and intervention have a positive influence on maintaining sobriety [13]. Nevertheless, the reality is insufficient compared to physiological research [14], and no sleep scale exists for patients with alcohol use disorder [15].
However, the assumption that insomnia disappears if one achieves sobriety without professional sleep intervention may increase the relapse of substance addiction [15]. As good sleep is an important indicator of recovery from addiction [16], professional management is necessary. To apply such sleep interventions, the drinking history of the person with alcohol use disorder must be identified, and a scale for this purpose must be used. Nonetheless, in clinical practice, sleep problems are evaluated using general sleep scales, and sleep disorders are identified by asking questions about drinking [14,16]. These clinical situations may make it difficult to evaluate sleep disorders in patients with alcohol use disorder. As these patients often use alcohol as a substitute for hypnotics [1] a sleeping scale that reflects this information is required. In the mental health field, there are few self-reported sleep scales for people with substance use disorders due to a lack of interest from researchers and mental health professionals in the process of developing these scales [16]. In South Korea, the situation is similar, and little research has been conducted to develop appropriate scales.
As sleep disorders experienced by people with substance use disorders have characteristics different from those of the general population, it is necessary to evaluate the substances taken, detoxification experience, and sleep hygiene in inpatient facilities or at home. Moreover, alcohol consumption should be measured during the pre- and post-evaluation of sleep intervention. For this purpose, standardized and validated scales are necessary.
Substance Use Sleep Scale (SUSS) is a tool for measuring sleep first developed by Neale et al. [16] for people with substance use disorders. It was developed using a mixed methodology of quantitative and qualitative research in the United Kingdom and has the advantage of being a diverse population in substance use disorders. An initial set of 30 items was developed through a process of validation, leading to a final set of 23 items with two sub-factors: “Mind and body sleep problems” and “Substance-related sleep problems.” The SUSS has the advantage of allowing individuals with alcohol use disorders to monitor their own sleep and allows treatment providers and researchers to use the scale to inform interventions and sleep outcome measures [16]. Therefore, this study aims to verify the validity and reliability of the Korean version of the Substance Use Sleep Scale (SUSS-K), developed by Neale et al. [16] for people with substance use disorders appropriate for the domestic situation.
METHODS
Design
This methodological study translated the SSUS, developed by Neale et al. [16] into Korean and evaluated its validity and reliability.
Participants
The study participants were inpatients in special alcohol treatment wards of three psychiatric mental hospitals and outpatients in five community addiction management locations in five cities.
The inclusion criteria for selecting participants were as follows.
We included those aged 19-65 years who had been diagnosed with alcohol use disorder by psychiatrists according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, and those who did not have withdrawal symptoms at the time of data collection and whose symptoms were sufficiently stable for attending psychiatrists to determine whether they could respond to the survey. Those who understood the purpose of this study and agreed to participate were selected.
In this study, the number of participants in the scale development study ranged from 115 to 230 based on the standard [17] that 5-10 participants were needed per item of the measurement scale for exploratory factor analysis (EFA). Subsequently, for confirmatory factor analysis (CFA), based on Bryant and Yarnold’s [18] recommendation that the sample size be the number of items multiplied by 5-10, a minimum of 115 participants was required. Therefore, this study requires at least 230 participants.
Measurements
SUSS-K
The SUSS developed by Neale et al. [16] to investigate sleep disorders of participants with substance use disorders, was used with permission from the original author. This scale has a total of 23 questions scored as “yes” or “no.” The overall Cronbach’s α value of the scale at the time of development was 0.88. The Cronbach’s α value for each item was 0.86 for mental and physical sleep problems and 0.79 for sleep problems related to substance use.
Initially, we obtained permission to use the scale from the author who developed it, and the author sent us guidelines on the scale translation process from King’s College London. The guidelines are referenced upon request from the original author. The author requested that we perform the following procedures: primary translation, expert review, back translation, expert review of the back translation, participation of at least five patients in the preliminary investigation, and final proofreading. Accordingly, we conducted the present study.
The primary translation was conducted in Korean and English by a translation expert fluent in English, whose native language was Korean. Afterwards, two psychiatric nursing professors fluent in English and a primary translator performed the back-translation. In this process, the consistency and accuracy of the original and translated sentences were confirmed, and changes were made to suit the Korean status, considering the differences in readability and cultural meaning.
Afterwards, corrections were made by back-translating to foreigners who had no knowledge of the scale, whose native language was English, and who were fluent in Korean. The scale was translated into Korean and underwent a cognitive evaluation process with one psychiatrist, four psychiatric mental health nurses, and two psychiatric mental health social workers. In this process, the word “alcohol” in the question “I have needed alcohol or drugs to get out of bed” was modified to “hangover liquor (haejangsul in Korean).” This word reflects the opinion that psychiatric mental health professionals should use the language of the participants. Each item on the scale was read aloud, and modifications were made considering reading comprehension, readability, and vocabulary for those with alcohol use disorder. After completion of this process, 23 questions were answered.
Content validity
To verify content validity, eight experts were selected: one psychiatrist, one psychiatric mental health nursing professor, three psychiatric mental health nurses who worked in an alcohol ward with certification as addiction specialist nurses, and two psychiatric mental health social workers. The content validity index (CVI), scale CVI/universal agreement, and item CVI were calculated. The CVI was measured on a 4-point scale, with each item ranging from 1 point for very unsuitable to 4 points for very suitable. According to Lynn’s guidelines [19], a CVI score of 0.80 or higher was set as the criterion for securing content validity. In this process, items 16 and 18 of the SSUS, which were less than 0.08, were deleted, reflecting the expert’s opinion that the questions asked about the use of street drugs were not appropriate for the Korean situation.
Experts suggested that because street drugs are interpreted as street drugs and are illegal in Korea, they would not be able to obtain accurate answers from the participants. Instead of these two questions, the following questions were added: “I feel like I am not asleep when I wake up in the morning” and “If I wake up during the night, I cannot go back to sleep.”
Therefore, the final number of questions remained unchanged at 23, but the contents of two questions were revised and supplemented. The 23 questions finally selected reflected the opinion that the 4-point Likert response scale was more appropriate than the yes/no response of the original scale. The answer format was changed to the specific number of times: “Not applicable,” “Sometimes (1-2 times a week),” “Often (3 times a week),” and “Always (every day)”; scores ranged from 1 to 4. This has been revised and supplemented so that patients with alcohol use disorder can respond more easily. We reported this change in the response format to the original author and confirmed that there were no problems.
Data collection
Data were collected from June 2023 to November 2023. We visited three psychiatric mental hospitals and five community addiction management centers in five cities. Questionnaires were distributed to patients who understood the purpose of the study and agreed to participate, and the completed questionnaires were collected.
Considering the minimum sample size of 230 for validation and the dropout rate, a total of 300 questionnaires were distributed. Of these, 292 questionnaires were collected as final data after excluding 8 questionnaires with untruthful or missing responses.
Therefore, 292 participants were randomly assigned using the SPSS program and classified into Data A (n=115) and Data B (n=177). For EFA and CFA, the required sample size was secured.
Ethical considerations
This study was approved by the Institutional Review Board of the researcher’s institution (approval number: JBNU 2023-05-022-001). All participants were informed that they could withdraw from the study if they did not want to participate at any point. Written informed consent was obtained from all participants.
Data analysis
Data analysis for this study was conducted in the following order using SPSS WIN (version 26.0, IBM Corp.) and AMOS 26.0 (IBM Corp.). The general characteristics of the participants were analyzed using frequencies, percentages, means, and standard deviations using descriptive statistics. Independent t-tests and chi-square tests were conducted to verify the homogeneity of the participants in EFA and CFA. A CFA was performed to verify the item analysis, EFA, reliability analysis, and model’s goodness of fit.
This study conducted item, exploratory factor, and CFA to verify construct validity. Prior to conducting the factor analysis, an investigation was conducted into the skewness, kurtosis, mean, and standard deviation of each item, as well as to examine whether it was normally distributed. Whether the correlation coefficient between the items and all items was less than 0.30 and the reliability coefficient Cronbach’s α value were checked, and deletion was considered [18,20]. To confirm the suitability of the factor analysis, Kaiser-Meyer-Olkin (KMO) and Bartlett’s tests for sphericity were performed.
The EFA used varimax rotation as a principal component analysis method to extract appropriate factors while maintaining independent correlations. The number of factors was determined as follows: eigenvalue of 1.00 or more, cumulative explanatory power of 60.0% or more, factor loading of 0.50 or more, and commonality of 0.40 or more [20]. To evaluate the goodness of fit of the research model, CFA was used to examine the root mean square error of approximation (RMSEA) and standardized root mean-square residual (SRMR), which were calculated using the goodness-of-fit index χ2 statistic (p), induction ratio, and residual, and the values of the incremental fit indices, such as the Tucker-Lewis index (TLI), incremental fit index (IFI), and comparative fit index (CFI) were examined.
To verify convergent validity, standards of standardized regression weight>0.50, construct reliability>0.70, and average variance extracted estimate>0.50 were applied [21]. Additionally, if the average variance extracted was greater than the square of the correlation coefficient, discriminant validity was considered to have discriminant validity [22].
In this study, reliability was verified by examining internal consistency based on the average correlation coefficient between each item within the scale, and Cronbach’s α was evaluated.
Test-retest reliability was analyzed by calculating the intraclass correlation coefficient (ICC) between the first and second administration of all items.
RESULTS
General characteristics
Examining the general characteristics of the participants in Data A for EFA, 106 (92.2%) were males, and the majority were 60-79 years old, 57 males (49.5%), and a mean age of 56.91 years. Examining the general characteristics of the participants in Data B for CFA, 153 (86.4%) were males and 24 (13.6%) females, with a mean age of 56.37 years.
As a result of the homogeneity verification conducted on EFA data A and CFA data B, there were no significant differences in any general characteristics.
Further details about the general characteristics are presented in Table 1.
Content validity
The CVI of the scale was calculated by a group of experts. For items with low scores, we tried to revise them by reflecting on the opinions of experts, but there were no items below 0.80. The overall CVI was 0.92, and out of 23 items, 13 (56%) had a CVI of 1.0, 10 (44%) had a CVI of 0.8, and no items were deleted.
Construct validity
Item analysis
Item analysis was first conducted. As a result of analyzing a total of 23 items, the average score of the items ranged from 2.27 to 3.40, the skewness value ranged from -0.04 to 1.96, and the kurtosis value ranged from -1.64 to 2.81. It was confirmed that the items were normally distributed, and both skewness and kurtosis met the recommended standards (skewness ≤±2.0, kurtosis≤±7.0) [22].
Analysis of the corrected item-total correlation (ITC) coefficient showed that the correlation coefficients of all items were over 0.30, confirming internal consistency. Thus, 23 items were selected (Table 2).
EFA
EFA was conducted three times on the selected 23 items using principal component analysis with varimax rotation, using data from 115 randomly selected individuals. Before performing factor analysis, KMO and Bartlett’s sphericity tests were performed to determine whether the sample was appropriate.
The KMO values were all 0.89 (reference value 0.50), which is close to 1, and the results of Bartlett’s sphericity test showed that the first χ2=1,239.14 (p<0.001), the second χ2=11,927.73 (p<0.001), and the third χ2=1,239.16 (p<0.001), which was confirmed to be suitable for factor analysis. The number of factors was determined based on items with an eigenvalue of 1.0 or higher, an explanatory power of 60.0% or higher, a factor loading of 0.50 or higher, and a commonality of 0.40 or higher [20].
As a result of the first factor analysis, four factors with eigenvalues of 1.00 or more were extracted, and item 2, “I have wanted to sleep better,” which had a factor loading of less than 0.50, was deleted.
After excluding one item deleted from the first factor analysis, a second factor analysis was conducted. As a result, the factor loading of item 17 “I have drunk alcohol to help me sleep” and item 19 “I have woken up in the night and smoked tobacco” were less than 0.50, so the two item were deleted.
As a result of the third factor analysis, the number of factors in the elbow point of the scree diagram was confirmed to be 4, Cronbach’s α value was 0.91, unchanged. The explanatory loading was 0.50-0.80, commonality 0.38-0.78. As for the explanatory power of was 37.69% for factor 1, 10.22% for factor 2, 8.28% for factor 3, and 6.61% for factor 4, respectively. The cumulative explanatory power was 62.79%, which is higher than the standard value (Table 3).
The four factors extracted through EFA were named factor 1 “difficulty maintaining sleep,” factor 2 “difficulty in initiating sleep,” factor 3 “difficulty waking up,” and factor 4 “poor lifestyle habits,” respectively.
In factor 4, the item “I have woken up in the night and drunk alcohol” had a high factor loading of 0.55 in factor 1. After reading it repeatedly and understanding its meaning, the two researchers decided that it was appropriate to group it as factor 4 (Table 3).
CFA
To verify the construct validity of the scale for the 4 factors and 20 items identified through EFA, CFA was conducted using data from 177 people who did not overlap with the participants of the EFA.
Normality was satisfied with the item analysis, and the goodness-of-fit index of the model was evaluated using the maximum likelihood method based on four factors. Consequently, χ2=656.95 (p<0.001), degrees of freedom (df)=287, χ2/df=2.29, SRMR=0.06, RMSEA=0.062, the 90% confidence interval (CI) of RMSEA is 0.058-0.070, and the model was confirmed to be suitable for all except the χ2 statistic, which is sensitive to the number of samples. The incremental fit index was also found to satisfy the criteria with IFI=0.90, TLI=0.92, and CFI=0.92 (Table 4).
Convergent validity and discriminant validity
The convergent and discriminant validity of the 20 items was verified using a multi-trait multi-item matrix analysis. As a result, the correlation coefficients between the 20 items and the subscales to which each question belonged ranged from 0.50 to 0.77, and were all above 0.40, indicating that the convergent validity of the items was 100% successful. Regarding the discriminant validity of the items, the correlation coefficient with other subscales to which each item does not belong ranges from 0.22 to 0.68, and there are no items with a value greater than the correlation coefficient compared with the factor to which each item belongs.
Simultaneously, if there was a difference between the correlation coefficient of each item and its subscale, the correlation coefficient with other subscales, and the correlation coefficient with the subscale, the discriminant validity of the item was considered established. The correlation coefficient with others and the Cronbach’s α coefficient of each subscale was higher than the correlation coefficients between the other subscales. Therefore, the characteristics of each subscale were distinct, and the item discriminant validity success rate was 100% (Table 5).
Reliability analysis
To verify the reliability and homogeneity of the scale, the ITC and the internal consistency Cronbach’s α coefficients were evaluated. The ITC value was 0.32-0.67, which was higher than 0.30, satisfying the standard, and it was confirmed that there was a positive correlation between all items.
The internal consistency Cronbach’s α for all 20 items was 0.91. The internal consistency Cronbach’s α for each subfactor were 0.83, 0.83, and 0.81, and that for the fourth factor was 0.74 (Table 4). This value is above the standard value of 0.70 [23].
Test-retest reliability was assessed to verify the stability of SUSS-K. The test-retest Cronbach’s α for the 18 subjects who participated in the first survey was 0.95, and the Cronbach’s α for each factor were 0.88, 0.88, 0.91, and 0.79 for the factor 4. The final ICC value was 0.91 (95% CI, 0.895 to 0.936; p<0.001), which confirmed the stability of SUSS-K.
Selection of final item
This scale was developed to identify sleep disorders in patients with alcohol use disorder. Through the validity and reliability verification process, a 4-point Likert scale instrument consisting of 20 items on four factors was confirmed. The items were structured according to the following sub-factors: difficulty maintaining sleep, difficulty initiating sleep, difficulty waking up, and poor lifestyle habits.
The measurement range of this scale is a Likert scale ranging from 1 (never) to 4 (always). The total score ranged from 20 to 80 points, with higher scores indicating more severe sleep problems.
DISCUSSION
This study aimed to validate the Korean version of the SUSS originally developed by Neale et al. [16] In order to ensure the content validity of the scale, this study was conducted according to the scale process guide provided by King’s College London, where the original scale developer belongs.
The construct validity was verified through exploratory and confirmatory factor analyses. The factor structure of this scale was confirmed at the time of development; therefore, it was appropriate to check its validity using CFA rather than EFA [22]. Nevertheless, as the SSUS was developed in England, validity verification was conducted to confirm that it is a suitable scale for Korean participants with alcohol use disorder. EFA and CFA were conducted. As a result of EFA of the 23 items extracted in the content validity process, the final 20 items were selected, and the cumulative explanation rate of the scale was 62.79%, ensuring satisfactory explanatory power despite the small number of items [20].
Unlike Neale et al. [16] who divided 23 items into two sub-factors at the time of development, this study classified them into four sub-factors. In the original scale, two subfactors were derived: “mind and body sleep problems” and “substance-related sleep problems.” However, the scale developed in this study reflects the passage of time. Reflecting on this, the researchers named the subfactors as difficulty maintaining sleep, difficulty initiating sleep, difficulty waking up, and poor lifestyle habits.
The difference between the results of this study and the number of factors in the SSUS suggests that there are differences between Eastern and Western languages and cultures. Although the original scale of the SSUS included participants with drug use disorders, this study was conducted only on al-cohol use disorders; therefore, there may be differences by factor. In addition, it can be inferred that the change in the response method of the original scale from a yes/no format to a 4-point Likert scale affected the number of factors; however, this needs to be confirmed through repeated research in the future.
Ultimately, three items from the original scale were excluded. Looking at this, item 2 “I have wanted to sleep better,” item 17 “I have drunk alcohol to help me sleep,” and item 19 “I have woken up in the night and smoked tobacco” were deleted. Item 2 is assumed to have been eliminated because it appeared to be hopeful content about sleep rather than an evaluation of sleep problems. Furthermore, the deleted items 17 and 19 were interpreted as being eliminated because they were classified as having similar content to question 18, “I have woken up in the night and drunk alcohol.” As repeated measurements of similar items may make it difficult to understand their content, this scale has improved readability [24] and is believed to be more effective when applied to participants with alcohol use disorder.
In terms of general characteristics, 33 female patients participated, a lower participation rate than male patients. Historically, sleep studies on patients with alcohol use disorders have been conducted mainly on men, with only a few of women participating [2,25]. In Korea, the prevalence rate is 3.3 times higher in men than in women, and many women do not receive practical help for addiction problems and treatment due to prejudice and stigma against female alcohol use disorders [8,26]. Most female patients from the five institutions where this study was conducted participated, but it can be seen as reflecting the reality of Korea. Fortunately, the original SUSS was reported to have demonstrated reliable and robust validity, with a 36% female participant population, and to be applicable to a wide range of populations with substance use disorders [15].
“Difficulty maintaining sleep” of the sleep disorder measurement scale extracted was the first factor and consisted of 5 items and showed the highest explanatory power at 37.7%. This factor consisted of items on fear, nightmares, pain, and distress during sleep. When alcohol is taken for insomnia, REM sleep increases, frequent awakenings occur, and sleep efficiency decreases [27-29]. Thus, as alcohol increases the occurrence of movement disorders and parasomnias during sleep [7], it is related to the contents of the extracted items of factor 1. Taking moderate doses of alcohol suppresses melatonin secretion during sleep; in particular, participants with alcohol use disorder have difficulty sleeping, as melatonin levels are not maintained until approximately 2 AM [7,30]. Therefore, this factor represents the reality of physiological problems associated with alcohol consumption and sleep. Sleep maintenance insomnia is the most common symptom in substance-addict-ed patients, accounting for more than 50% [11,12], so the first factor was named “difficulty maintaining sleep.”
The second factor, “difficulty in initiating sleep,” comprised six items. This factor consisted of items asking about difficulty sleeping due to uncontrollable thoughts such as anxiety when trying to fall asleep, with an explanatory power of 10.2%. Prolonged exposure to alcohol results in the loss of the sleep-inducing effect of alcohol is lost [1]. Additionally, those who continuously drink alcohol have longer sleep onset times and reduced total sleep time [27]. Chronic alcohol use damages the brain’s reward system, causing sleep problems that lead to excessive negative emotions such as anxiety and dysphoria [31]. Depression and anxiety caused by insomnia frequently appear in patients with substance use disorders and increase the risk of sleep problems [11,13]. This was connected to the derivation of items about negative emotions such as anxiety, anger, and guilt in the second factor. Additionally, the item “I have woken up with a hangover or drunk” was also included. This suggests that as tolerance increases with alcohol use, the initial drowsiness-inducing effect is reduced and a larger amount of alcohol is required, which may worsen sleep problems. This factor included the anxiety that people with alcohol use disorder drink to induce sleep or that they may drink again if they cannot fall asleep, so it was named “difficulty initiating sleep.”
The third factor was “difficulty waking up” and comprised five items, with an explanatory power of 8.3%. Factor 3 consisted of items about feeling tired when waking up in the morning and not feeling like you are sleeping. Sleep deprivation in alcohol patients can increase impulsivity and decrease attention [32]. Alcohol tends to cause sleep disruption in the latter half of sleep, and not drinking alcohol within 6 hours before bedtime helps with good sleep [9]. Nonetheless, if these problems are not resolved, neurobiological pathways in the brain repeat a continuous vicious cycle of alcohol use and sleep disorders [31]. The third factor was named “difficulty waking up” to reflect sleep problems in patients with alcohol use disorder.
The fourth factor was named “poor lifestyle habits” and comprised four items with an explanatory power of 6.6%. Factor 4 comprised items such as taking alcohol after waking up in the early morning or after waking up and being too tired to work during the day. Korea may be socially and culturally tolerant of alcohol; therefore, the rate of alcohol use disorder will be increasing. The sociocultural phenomenon of hangover liquor consumption in Korea can be seen as a rationalization of drinking habits that increases its prevalence. Deterioration in sleep quality due to alcohol consumption is the most common problem that increases daytime sleepiness [33]. Taking a nap due to daytime sleepiness can cause insomnia at night; therefore, lifestyle changes are necessary. However, the number of awakenings increases as withdrawal and sobriety are repeated, and sleep problems become more severe 9 months after abstinence may worsen [34]. Therefore, sleep hygiene management is essential when living at home. Fortunately, insomnia can be alleviated with sobriety and improvement of other lifestyle factors [28], suggesting the need to develop interventions based on this. The fourth factor, “poor lifestyle habits,” suggests that there is a need to educate those who live at home after discharge or use rehabilitation centers about proper sleep hygiene and lifestyle habits.
In this study, CFA was conducted on 20 of the four derived factors to verify the construct validity of the SUSS-K. In this study, the Goodness of fit index results showed that it generally fit all criteria; therefore, the model was judged to be relatively appropriate. Through discriminant validity verification, it was confirmed that there was a high correlation between the items and the sub-factors and that the sub-factors were independent of each other. As a result of reliability verification, the Cronbach’s α value of the developed scale was found to be 0.74 to 0.83. Generally, the smaller the number of items, the lower is the Cronbach’s α value [35]. However, as a result of analyzing the reliability of this scale, Cronbach’s α for all items was 0.91, and that for each subdomain was 0.74-0.83, indicating high reliability 0.70, which is the standard suggested by Nunnally and Bernstein [36]. It was judged to be a reliable scale with good internal consistency.
This study had some limitations. The linguistic compatibility of each question was checked and corrected through a double translation method [17-18,21] according to the recommendation of the original scale researcher in this study, but there was a difference in the final number of items and the number of items. Owing to the number of sub-elements of the original scale, it must be reconfirmed through a follow-up study. In addition, SUSS was developed for alcohol and drug use disorders, but this study collected data only for alcohol use disorders. Therefore, it is necessary to be cautious when applying this method to participants with substance use disorders, and further research that includes participants with drug use disorders is necessary.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
