Alcohol dependence and the psychological factors leading to a relapse: a hospital-based study in Vietnam

Purpose – To study the situation of alcohol relapse and to investigate the relationship between psychological factors and alcohol relapse in persons with alcohol dependence in Thai Nguyen hospitals, Vietnam. Design/methodology/approach – A correlation study was conducted among 110 patients. Data were collected through structured interviews and were analyzed using descriptive statistics and Spearman ’ s correlation coefficient ( r s ). Findings – More than two-thirds of the participants were found to relapse more than once ( X 5 2.04, SD 5 0.86). Positive outcome expectancies, cravings, negative emotional states, and maladaptive coping were positively associated with relapse ( r s 5 0.550, 0.522, 0.497; p 5 0.000 and r s 5 0.217, p < 0.05, respectively). While, motivation to change with three subscales had a negative correlation to relapse including recognition (r s 5 (cid:1) 0.199, p < 0.05), ambivalence ( r s 5 (cid:1) 0.331, p 5 0.000), and taking steps ( r s 5 (cid:1) 0.606, p 5 0.000). Adaptive coping, self-efficacy, and social support were also found to be negatively correlated to relapse ( r s 5 (cid:1) 0.535, (cid:1) 0.499, (cid:1) 0.338; p 5 0.000, respectively). However, negative outcome expectancies ( r s 5 (cid:1) 0.024, p 5 0.805) and positive emotional states ( r s 5 0.081, p 5 0.399) were not significantly related to relapse. Originality/value – This is the first study in Vietnam which investigated the relationship between psychological factors and alcohol relapse in individuals with alcohol dependence. correlational design was used to investigate the relationship between self-efficacy, outcome expectancies, cravings, motivation, coping, emotional states, social support and alcohol relapse.


Introduction
In western countries, alcohol is consumed because of its quality and in order to appreciate and enjoy the beverage [1]. In the case of Vietnam, drinking is also socially accepted as evidenced by the popular catch-phrase "guests coming home were invited to tea or alcohol" [2]. Drinking is also seen as an opportunity to display masculinity as seen in the Vietnamese quote: " A man without alcohol is like a flag without wind" [3]. Needless to say, alcohol consumption is a normal part of daily life in Vietnam.
A study of alcohol consumption showed that 80% of Vietnamese males were reported as alcohol drinkers, and amongst the 80%, 40% were considered heavy drinkers [4]. This is a large burden to Vietnam both in terms of the cost of medical treatment and the social issues created by a heavy drinking culture [5]. In Vietnam, 7.3% of all deaths were estimated as being related to alcohol consumption in 2016 [6]. In the same year, nearly 10% of men aged 50 to 69 years old were reported to have died of alcohol-related liver cancer [7].
Despite preventive efforts, many people relapsed after a period of abstinence from alcohol drinking [8]. Alcohol relapse is a term used to describe the return to previous levels of drinking and is a complex phenomenon [9]. Various factors are correlated to alcohol relapse such as self-efficacy [10], outcome expectancies [11], coping strategies [12], emotional states [13], motivation [14], cravings [15] and social support [16].
After many decades of study, alcohol relapse is a common occurrence during the recovery process. This is found to be consistent in a study in Vietnam, where it was reported that 62.8% of alcohol-dependent patients had relapsed after treatment completion [17]. This is a worrying statistic as relapse causes various problems related to individual health, family conflict, public health and social problems [18].
In Vietnam, Thai Nguyen province has a high prevalence of alcohol consumption [19]. Residents believe that alcohol is an indispensable part of socializing and refusing to drink is even regarded as rude or worse [20]. In 2010, 19.7% of problem drinkers displayed alcohol dependence symptoms [21]. This number is higher compared to other provinces [22,23]. Further, statistics from Thai Nguyen Psychiatric Hospital showed that the number of patients hospitalized due to alcohol dependence occupied nearly 42.6% of total patient beds. Furthermore, most patients (70%) unfortunately returned to drinking after treatment completion [24].
Understanding factors related to relapse allows clinicians to develop effective relapse prevention strategies [25]. Although there have been a number of studies conducted on the topic, factors correlating to relapse were varied and not consistent among different groups of peoples and cultures [26][27][28]. Conducting a context-specific study could provide information on the local relapse situation. There is one study in Vietnam that has explored alcohol relapse but it did not evaluate the association between alcohol relapse and its related factors [17].
Therefore, this study was designed to explore the relapse situation and to investigate the relationship between alcohol relapse and the several psychological factors amongst the Vietnamese with alcohol dependence in Thai Nguyen province.

Research design
A correlational design was used to investigate the relationship between self-efficacy, outcome expectancies, cravings, motivation, coping, emotional states, social support and alcohol relapse.

Settings
Thai Nguyen is a mountainous province in the northern region of Vietnam. The natural land area is more than 3,500 square kilometers with a population of around 1.2 million [29]. There are eight provincial hospitals and one national hospital. However, there is no specific hospital for the treatment of alcohol addiction. Patients with alcohol dependence are treated at general psychiatric hospitals or general medical hospitals.
Thai Nguyen National Hospital is a national general hospital with 1200 beds offering a wide spectrum of services. It has a psychiatric department with 24 beds that provides treatment for people with psychiatric disorders and alcohol dependence.

Alcohol relapse and psychological factors
There are also three general provincial hospitals that provide general services for local people including Thai Nguyen A Hospital (510 beds), Thai Nguyen C Hospital (510 beds), and Thai Nguyen Gang Thep Hospital (350 beds). In these hospitals, the General Medical Department provides treatment for persons with alcohol dependence. Additionally, this department also provides treatment for patients with liver and gastrointestinal diseases due to alcohol consumption.
There are five more specialist hospitals in the area including a Psychiatric hospital, a Tuberculosis and Lung Disease hospital, a Traditional Medicine hospital, an Orthopedic and Rehabilitation hospital, and an Eye hospital. Of these, Thai Nguyen Psychiatric Hospital (150 beds) is the only hospital that caters to alcohol-dependent patients. The Male Department (50 beds) is the department providing treatment for alcohol dependence, as well as other psychiatric disorders such as schizophrenia and depression.
Although there are hospitals for patients with alcohol dependence, the current treatment in both psychiatric and general hospitals are heavily biased toward treatment such as detoxification, managing alcohol withdrawal symptoms, electrolyte rehydration and nutrition. It is noticeable that alcohol-dependent patients in the general hospital are transferred to the psychiatric hospital if they have more severe psychotic disorders such as delusions, hallucinations or agitation.
Similarly, nursing interventions provided to patients with alcohol dependence are limited to medical technological interventions such as injection, infusion and physical examination. Relapse prevention training and nursing education are rarely provided. There is also no follow-up service. As a result, most of the alcohol-dependent patients have no guidance to avoid relapse after discharge [20].

Participants
A purposive sampling method was used. Patients with alcohol dependence were enrolled following a presentation at the selected hospitals. Participants aged more than 60-years-old were screened for their mental status using the 6-item cognitive impairment test [30].
The sample size was calculated using a power analysis with an alpha of 0.05, power of 0.80 and the effect size of 0.28. To account for the missing data, we included an additional 10% increase in subjects. Therefore, the number of subjects required for this study totaled 110 patients (Table I). In each hospital, there is only one department which admits patients with alcohol dependence and the researcher purposely chose these departments from which to recruit the participants.
The participants were selected using the following inclusion criteria: ( To minimize the collection bias, the researchers purposely selected the participants from the same population and used the same inclusion criteria in the five hospitals.

Instruments
A General Information Form was used to collect the demographic characteristics of the participants.
Alcohol relapse was defined as the number of patients with a re-emergence of alcohol dependence syndrome as per the International Classification of Disease Tenth Version (ICD-10) diagnostic criteria after a period of abstinence for at least one month in the last 12 months [31]. The measurement of alcohol relapse was developed by researchers. Cohen's Kappa (κ) was used to test for reliability and indicated a substantial agreement (κ 5 0.68) [32].
Drinking Refusal Self-Efficacy Questionnaire [21] with 20 items was used to measure the ability to refuse alcohol in three different situations: social pressure to drink, opportunistic drinking and emotional relief gained from the effects of alcohol. Each item was rated on a 4-point Likert scale. The higher the score, the better is the self-efficacy. The Cronbach's alpha was 0.86.
Drinking Expectancy Questionnaire (DEQ) [33] listing with 37 items was used to measure outcome expectancies. There were two subscales: positive and negative outcome expectancies. Each item was assessed on a 5-point Likert scale. Higher scores indicated more positive outcome expectancies. The Cronbach's alpha of the DEQ's subscales was 0.88 and 0.74.
Penn Alcohol Craving Scale [34] with 5 questions was used to measure alcohol cravings. This scale measured the frequency, intensity, duration, resistance and the average level of alcohol craving. Each question was scaled from 0 to 6. Higher scores indicated more frequent and more severe cravings. The Cronbach's alpha was 0.98.
Stage of Change Readiness and Treatment Eagerness Scale version 8 for alcohol (SOCRATES-8A) [35] with 19 items was used to measure motivation to change drinking behaviors. It comprised three subscales: recognition; ambivalence and taking positive steps. Each item was scored using a 5-point Likert scale. The higher the score, the better is the motivation. The Cronbach's alpha of three subscales of SOCRATES-8A was 0.77, 0.76 and 0.89.
Coping Behaviors Inventory (CBI) [36] including 36 items was used to measure coping strategies. There were two subscales: adaptive and maladaptive coping. Each item was rated on a 4-point Likert scale. The higher score indicated a higher frequency of use. The Cronbach's alpha of the CBI's subscales was 0.79 and 0.80.
Positive Affect and Negative Affect Schedule [37] was used to identify the emotions that caused patients to be more likely to drink. This scale consisted of 10 items of Positive Affect (PA) and 10 items of Negative Affect (NA). The responses used a 5-point Likert scale. The higher scores indicated a higher effect of PA or NA on drinking. The Cronbach's alpha of PA and NA was 0.74 and 0.83.
Multidimensional Scale of Perceived Social Support [38] including 12 items was used to assess social support from friends, family and significant others. Items were measured on a 7-point Likert scale. The higher the scores, the stronger is the support. The Cronbach's alpha was 0.83.

Alcohol relapse and psychological factors
Ethical consideration This study was submitted for ethical approval to the Institutional Review Board (IRB) of Hanoi University of Public Health, Vietnam. Approval was granted on January 31, 2019. The IRB code was 019-010/DD-YTCC.

Data collection
Data were collected from March 1, 2019 to April 1, 2019, using structured questionnaires and face-to-face interviews. Before collecting data, the researcher gathered permission to collect data from the Directors of each hospital in Thai Nguyen, Vietnam. After that, nurses were asked to identify patients with alcohol dependence. Following that, selected patients were asked if they would be willing to participate in this study and those who met the inclusion criteria were recruited to join the study. If patients consented, the researcher would proceed to explain all steps and human rights protection, and subsequently interviewed the participants using all the selected questionnaires. The duration of the interview was about 30-45 minutes.

Data analysis
Data were analyzed using the Statistical Package for Social Sciences, version 17.0 (SPSS 17.0). Descriptive statistics and Spearman's correlation coefficient were used.

Results
Characteristics of the participants All participants were males, with a mean age of 49.08 (SD 5 0.76). Most of them (75.5%) were from Thai Nguyen province and 98.2% were married. Nearly two-thirds (64.5%) had attended secondary school. Approximately 65% were farmers, and 62.7% had a monthly income greater than the minimum standard of daily living. Participant's mean age of first alcoholic drink was 17.9 years old (SD 5 2.4), with 11.4 years (SD 5 5.2) mean duration of alcohol dependence. Of all participants, 77.3% of participants reported a positive family history of drinking (Table II).

Characteristics of dependent and independent variables
Alcohol relapse was the dependent variable. The number of alcohol relapses ranged from 1 to 4 with a mean (SD) of 2.04 (0.86). In the study, 34 participants (30.9%) reported relapsing just once in the previous 12 months. Notably, 69.1% of the participants relapsed more than once. In particular, 42 (38.2%) relapsed twice, 30 (27.3%) thrice, while 4 (3.6%) participants relapsed four times (Table III).
Regarding independent variables, self-efficacy was at a moderate level with a mean (SD) of 46.79 (4.54). The mean of positive and negative outcome expectancies was 72.25 (9.25) and 46.48 (6.58), respectively. Craving was at a moderate level with a mean of 17.10 (7.42). Motivation to change was at a low level. The mean of recognition, ambivalence, and taking positive steps subscales was 22.75 (3.69), 12.80 (2.82), and 25.84 (5.31), respectively. While the mean of adaptive coping and maladaptive coping was 14.68 (4.12) and 15.42 (5.14), respectively. The mean of positive and negative emotional states was 17.00 (3.27) and 24.74 (4.56), respectively. Social support was at a moderate level with a mean of 49.90 (6.42) as shown in Table IV.

Relationship between independent variables and alcohol relapse
There was a positive relationship between alcohol relapse and positive outcome expectancies, cravings, negative emotional states and maladaptive coping (r s 5 0.550, 0.522, 0.497; p 5 0.000 and r s 5 0.217, p < 0.05). Motivation to change drinking behavior within the three Alcohol relapse and psychological factors subscales had a negative relationship with alcohol relapse at a low to a high level: recognition (r s 5 À0.199, p < 0.05), ambivalence (r s 5 À0.331, p 5 0.000), and taking positive steps (r s 5 À0.606, p 5 0.000). Similarly, adaptive coping, self-efficacy and social support were negatively correlated to alcohol relapse (r s 5 À0.535, À0.499, À0.338; p 5 0.000). In contrast, negative outcome expectancies (r s 5 À0.024, p 5 0.805) and positive emotional states (r s 5 0.081, p 5 0.399) were not significantly related to alcohol relapse (Table V)

Characteristics of participants
The majority of participants (89%) were 40-to 59-years-old. This indicates that alcohol dependence and relapse was more prevalent among individuals in the late-adult group. This finding is supported by a population survey in Vietnam which reported drinking to be common among those between the age of 46-54 years old [39]. A possible explanation for the large proportion of male alcohol consumers can be attributed to the Vietnamese culture where women are not expected to consume alcohol. Furthermore, there is a high expectation for men to drink with a common belief that a man who does not drink is considered a coward [3].
Most participants (98.2%) were married and 62.7% reported a monthly income greater than the minimum income for daily living. This could mean that the patients were not living in poverty, were financially independent and were more able to purchase alcohol. These findings were in accordance with a study in Vietnam which reported that the majority of the drinkers were currently married and lived independently within their community [40].
Most participants (76.4%) started to consume alcohol in their youth (16-20-years-old). The mean duration of alcohol dependence was 11.4 years, where 60.9% reported having experienced alcohol dependence for 5-10 years. These findings were supported by previous Vietnamese researchers who reported that most drinkers had been dependent for more than 5 years [41].
It was also observed that 77.3% of participants had a family member who also drank alcohol. This finding was congruent with other studies that indicated a positive history of alcohol use among people with alcohol relapse [42,43].

The situation with alcohol relapse
The findings showed that more than two-thirds of the participants (69.1%) had experienced more than one relapse in the last 12 months. Previous studies indicate that the greater the number of alcohol relapses, the more the severity of any relapse situation [42,44].  [17]. The high rate of alcohol relapse could be explained by the following reasons. Firstly, in Thai Nguyen, no alcohol restriction measure has been imposed. Alcohol is easy to buy in street stalls, through vendors, and at restaurants. Further, the cost of alcohol is comparatively low. People believe that drinking is commonplace, and alcohol plays an important role in Vietnamese society. As stated earlier, there is still a lack of hospital alcohol addiction treatment. Therefore, alcohol-dependent individuals have little help or skills in knowing how to avoid a relapse [20].

The relationship between independent variables and alcohol relapse
The findings showed that of the twelve factors hypothesized in correlating with alcohol relapse, ten factors showed a significant relationship at the alpha level of 0.05.
At first, positive outcome expectancies were considered as the most important contributor with the strongest association to alcohol relapse (r s 5 0.550, p 5 0.000). The more individuals that valued the outcome of drinking, the more likely it was for them to engage in drinking and increase the probability of relapse [45]. Furthermore, it is no surprise that Vietnamese people have a strong belief in the positive effects of drinking. In the countryside, alcohol is commonly consumed during the new year period and at weddings and anniversaries [3] thereby creating limited awareness of alcohol's adverse effects. This finding was similar to a previous Japanese study which reported that high positive alcohol outcome expectancies significantly predicted a worse treatment outcome when eventually suffering from alcohol dependence [46].
Adaptive coping was the second main contributor to alcohol relapse (r s 5 À0.535, p 5 0.000). The reason is that individuals with adaptive coping had active and effective coping strategies that increased the ability to deal with stressful situations including the risk of alcohol reconsumption that, in turn, decreased the probability of relapse [47]. It is noteworthy that Vietnamese drinkers with alcohol dependence had to face the pressure to drink when they engaged in social events [3,20]. Therefore, the use of adaptive coping strategies would significantly help them to avoid alcohol relapse.
Also, there was a positive relationship between maladaptive coping and alcohol relapse in this study (r s 5 0.217, p < 0.05). The findings of coping strategies are seen as parallel with previous studies that found individuals who had abstained from alcohol tended to use a greater number of effective coping strategies than those who had relapsed [26,44].
Craving was the third factor that contributed to alcohol relapse (r s 5 0.522, p 5 0.000). Alcohol craving is very difficult to stop or avoid and can easily cause an unplanned return to alcohol consumption [48]. Further, most alcoholics may never completely avoid experiencing craving, even during the treatment period [49]. This finding is supported by a study conducted in Vietnam which reported that craving was the most common reason that caused relapse after one year of discharge [17]. Other studies also indicated that there was a positive relationship between craving and alcohol relapse [15,50].
Self-efficacy with a moderate correlation to alcohol relapse (r s 5 À0.499, p 5 0.000) was the fourth contributor. It indicated that the higher the level of drinking refusal self-efficacy, the lower is the probability of alcohol relapse. A possible explanation is that alcoholdependent individuals who have low levels of self-confidence and were not successful at abstinence also had a historically higher level of alcohol abuse. After a failed first attempt, their drinking habit often continued to a full relapse [25]. This finding is similar to a study conducted in Malaysia which highlighted that self-efficacy had a negative association with relapse (r 5 À0.790, p < 0.05) [51].
The fifth contributor to alcohol relapse among Vietnamese alcohol-dependents was negative emotional states (r s 5 0.497, p 5 0.000). One possible explanation here is that individuals could have had previous experiences where alcohol had helped relieve their negative mood states. Because of the long duration of alcohol dependence, the relief of negative emotional states by drinking was formed. Therefore, alcohol that was consumed as a relief from negative memories proved to be a hard habit to break [52]. Also, negative emotional states such as anger were significantly related to decreased self-control of individuals that might result in maladaptive behaviors including alcohol consumption [53]. This finding was supported by a study conducted in Vietnam which indicated that depression was one of the reasons for alcohol relapse [17]. Similarly, other researchers reported that depression and anxiety symptoms had a significant correlation to drinking with a Pearson's correlation of 0.35 and 0.39 (p < 0.05) [54].
Social support was another contributor to alcohol relapse (r s 5 À0.338, p 5 0.000). The reason is that high social support has empowered individuals to overcome the adverse effects of stress and reduce exposure to risk situations of drinking that would help decrease the probability of relapse [55]. Furthermore, individuals with high levels of social support perceived that they might receive support from their family, friends, and others and were more able to cope with negative life stressors [56]. This finding was supported by a study in China that found a positive relationship between social support and resilience (r 5 0.484, p < 0.01) [16].
Regarding motivation, taking positive steps had a high negative correlation with alcohol relapse (r s 5 À0.606, p 5 0.000), followed by ambivalence and recognition of habits and patterns (r s 5 À0.331, p 5 0.000, and r s 5 À0.199; p < 0.05, respectively). The findings reveal that increased awareness of drinking problems and taking part in more activities that included engaging in the change of one's drinking behavior lengthened periods of abstinence. It also suggested that motivation to change is a critical factor for the success of alcohol abstinence. These findings were consistent with a previous study that reported that the higher the score of motivation to change, the better the treatment outcome of alcohol addiction [14].
Contrary to the hypotheses, negative outcome expectancies were not significantly related to alcohol relapse (r s 5 À0.024, p 5 0.805). In this instance, whether participants had low or high negative expectations toward the effects of drinking, there was still no difference in the rate of relapse. A possible explanation is that the participants in this study were always under peer pressure to drink. (77.3% reported a positive family history of drinking), so although participants recognized the harmful effects of alcohol consumption, they had to drink to gain respect, complacency, and not to lose face [57].
Similarly, positive emotional states had a nonsignificant association with alcohol relapse (r s 5 0.081, p 5 0.399). This means that although the participants used alcohol when they had positive emotions, these drinks were not related to their relapses. This may be because positive emotional states were not the leading cause of alcohol consumption among the participants in this study.

Recommendations
The researchers suggest some interventions for the nurses who provide services for persons with alcohol dependence in Thai Nguyen as below.
Alcohol education. Firstly, nurses should provide education that could help the patients to understand that relapse is a normal part of the recovery process. Also, the patients must be taught to recognize the warning signs of relapse such as cravings and negative emotional states. The nurse should then support patients to develop coping skills to prevent relapse.
Similarly, it is necessary to reframe the perception of alcohol outcome expectancies among patients. The objective is to reduce the high expectation of the outcomes of drinking and help clients to identify the adverse effects of alcohol, as well as the impact that it has on a person's health, family and society.

Alcohol relapse and psychological factors
In addition, nurses could provide education to enhance the motivation to change drinking behaviors by providing appropriate medical information related to current drinking habits and their health status. In this way, patients would be guided in exploring their experience, feelings and their capacity to change their drinking problems.
Skills training. At first, most participants showed a low level of self-efficacy in refusing to drink, especially when in social pressure situations. Therefore, the patients should be trained on how to say "no" to alcohol and develop strong internal resilience skills to stay away from alcohol.
Secondly, there was a high-frequency use of maladaptive coping when patients experienced the risk situation of relapse. It is suggested that nurses and doctors provide information on effective coping strategies during depressive or angry situations and give patients a chance to practice these strategies in the hospital. Also, the participants reported a moderate level of craving. Thus, the nurses should teach patients how to deal with the cravings as they occur by practicing meditation and exercise as suggested examples. Also, the healthcare system in Thai Nguyen should provide aftercare services for alcoholdependent individuals as alcohol addiction does not stop after the treatment period ends.

Conclusions
Positive outcome expectancies, adaptive coping and cravings are important determinants of relapse. Other variables such as negative emotional states, self-efficacy, motivation, maladaptive coping and social support were found to be significant contributors to relapse. The findings are useful in guiding healthcare providers to develop essential interventions to prevent relapse. It also suggests future experimental studies on relapse prevention in Vietnam.