Abstract
To date, no screening tests for gambling disorders have been adapted and validated in Central and Eastern Europe. The aim of this study is to adapt the Problem Gambling Severity Index (PGSI) and Lie/Bet questionnaire (Lie/Bet) and assess their psychometric properties once translated for use with the Polish population. A mixed sample (N = 300) was drawn from venues, social media, snowballing and treatment centers. PGSI had a higher coefficient of predictive power than Lie/Bet. However, differences between validated tests are not significant. Validation of screening tests of gambling disorders showed the necessity for verification of the scale of interpretation of results when conducting tests in Poland, changing cutoff values. The PGSI and Lie/Bet tests are short and easy to apply, they can be implemented in various types of institutions: for screening patients in primary health care facilities and for identifying comorbid gambling disorders in alcohol- and drug-dependence treatment facilities, and in social welfare centers.
Key words: Europe, gambling, Lie/Bet, Problem Gambling Severity Index, problem gambling, psychometric properties, screening tests
1. Introduction
The International Classification of Diseases and Related Health Problems, 11th Revision (ICD–11; World Health Organization, 2018) defines gambling disorder as
a pattern of persistent or recurrent gambling behavior, which may be online (i.e. over the internet) or offline, manifested by: (1) impaired control over gambling (e.g. onset, frequency, intensity, duration, termination, context); (2) increasing priority given to gambling to the extent that gambling takes precedence over other life interests and daily activities; and (3) continuation or escalation of gambling despite the occurrence of negative consequences. The behavior pattern is of sufficient severity to result in significant impairment in personal, family, social, educational, occupational or other important areas of functioning. The pattern of gambling behavior may be continuous or episodic and recurrent. The gambling behavior and other features are normally evident over a period of at least 12 months in order for a diagnosis to be assigned, although the required duration may be shortened if all diagnostic requirements are met and symptoms are severe (ICD–11 webpage, 2020).
People with gambling disorders usually start looking for help when their dependence is already advanced (Potenza et al., 2002). Almost three quarters (71%) of people with a pathological gambling habit have never sought professional help, nor used self-help groups. In the USA, fewer than 10% of people with gambling disorders decide to undergo treatment (Toneatto & Millar, 2004), whereas according to a study conducted in Ontario, Canada, 18% of problem and pathological gamblers looked for any form of treatment. More often people are looking for help once the problem is more advanced (Potenza et al., 2002). There is usually a five-year lapse between the appearance of the first signs of the disorder and the start of medical treatment (Evans & Delfabbro, 2005). Young adults tend to ask for help less often (Petry, 2005). Screening tests make the identification of people with the risk of gambling disorders possible. Their aim is to detect disorder at an early stage and to enable the start of a treatment that would prevent the appearance of more serious consequences of the disease in the future. Studies show that screening tests are effective when it comes to the identification of and a reduction in the prevalence of gambling disorders (Fortune & Goodie, 2010). Nevertheless, only 7% of general practitioners (GPs) had been doing screening tests for gambling disorders (Evans & Delfabbro, 2005).
To date, there has been no adaptation and validation of the screening tests for gambling disorders in Poland. Even so, the Problem Gambling Severity Index (PGSI) screening test has been used in studies of the general population to monitor the situation. There is a lack of procedures to screen for gambling disorders in Poland. The results of the 2014 survey show that in the last 12 months before the study, every third Pole aged 15 years old and older had gambled at least once (34.2%), while 7.1% gambled every day or a few times a week (CBOS Report, 2015). Compared to the 2012 results, the prevalence of gambling increased within two years by about 10%; however, the distribution of the popularity of particular gambling games has not changed. In Poland, the most popular gambling games are lotteries, followed by scratch cards, SMS lotteries and Electronic Gambling Machines (EGM) gaming (CBOS Report, 2012). In 2011, 3.7% of respondents obtained a minimum of 1 point on the scale of the PGSI, and 0.2% met the criteria for problem gambling. In 2014, 5.3% of Poles scored at least 1 point on the PGSI, and 0.7% indicated problem gambling (8 and more points). Based on the benchmark method, the number of pathological gamblers in 2014 was estimated as 27,955 people, which is 0.09% of the population over the age of 15 years (CBOS Report, 2012, 2015). The number of people treated for gambling disorders has been growing steadily for years. In 2015, there were 4.5 times more patients treated than in 2008 (4775 and 1050, respectively; National Health Fund, 2016).
The Problem Gambling Severity Index (PGSI) and Lie/Bet Questionnaire (Lie/Bet) are both based on the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM–IV; American Psychiatric Association, 1994) and Fourth Edition, Text Revision (DSM–IV–TR; American Psychiatric Association, 2000) diagnostic classification. However, despite the publication of a new edition of the DSM diagnostic classification (Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, DSM–5; American Psychiatric Association, 2013) and updated diagnostic criteria for gambling disorders, studies using the chosen screening test are still published. In the literature, there is a lack of articles where screening tests based on the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD–10; World Health Organization, 2004) diagnostic criteria were used. The two selected tests (PGSI and Lie/Bet) fulfill functions that are expected from screening tools: they are short, easy to understand by people of different intellectual levels and clearly formulated.
The PGSI was developed to assess the prevalence of the risk of the appearance of problem gambling in the general population, inter alia to reach people who are not covered by clinical studies and who have less severe gambling disorders. It has been demonstrated as most valid and reliable in measuring problem gambling prevalence in Western populations (Colasante et al., 2013; Ferris & Wynne, 2001). The test has nine questions about the control of playing, money spent on gambling, loss chasing, borrowing money, stress and anxiety connected with gambling, financial troubles and feeling guilty after playing. Zero points on the scale is interpreted as no gambling problem, 1–2 points indicate a low gambling problem, 3–7 points indicate a moderate gambling problem, and scores of 8 points or more indicate a serious gambling problem. The PGSI test is characterized by a high sensitivity of 78% compared to the diagnoses made by clinicians (clinical assessment interviews) and 83% compared to the DSM–IV criteria. The specificity of the PGSI test was 100%, compared to the clinical assessment interviews and the DSM–IV criteria (Ferris & Wynne, 2001).
The PGSI has been adapted and validated in many regions of the world. In the Middle East, in Persia, the PGSI showed excellent internal consistency (Cronbach’s coefficient α = .90; composite reliability = .91). The Persian PGSI score correlated with the score on the Hospital Anxiety Depression Scale, which assessed depression (r = .54, p < .01) and anxiety (r = .40, p < .01; Griffiths & Nazari, 2020). In Japan, the PGSI also had excellent internal consistency (Cronbach’s coefficient α = .89) and moderate test–retest reliability (So et al., 2019). In Europe, adaptation and validation of PGSI was made in Italy and in Spain. In Italy, Cronbach’s α estimated was .87 and was considered a good reliability. A strong convergent validity with the Lie/Bet questionnaire and with both depression and stress scale instruments was found (Colasante et al., 2013). In Spain, the PGSI showed satisfactory construct validity. The internal consistency (αordinal = .97), as well as its convergent validity with the DSM–IV scores (r = .77, p < .001), was good (Lopez-Gonzalez et al., 2018).
The Lie/Bet Questionnaire is a simple screening test. It has only two questions based on 10 diagnostic criteria of the pathological gambling from the DSM–IV classification. In the studies, specificity of the test reached more than 80% and its sensitivity was at a level of 100% (Fortune & Goodie, 2010; Johnson et al., 1997). The Lie/Bet test also has a high positive and negative predictive value (Colasante et al., 2013). It is recommended by general practitioners for conducting screening in a primary health care facility, because of its simplicity and good parameters (Evans & Delfabbro, 2005).
The Lie/Bet questionnaire is not as popular as a screening tool as is the PGSI. Its adaptation and validation were implemented in Norway, where it showed both high sensitivity and high specificity. The negative predictive value was also high, but the positive predictive value was comparatively lower (Götestam et al., 2004).
The aim of this study is to adapt the PGSI, a commonly used international instrument, and the Lie/Bet short test for quick screening, and to assess their psychometric properties when they are translated for use for the Polish population.
2. Method
2.1. The adaptation of screening tests
The adaptation was done in accordance with guidelines elaborated by Baeton et al. (2000) and included five stages: (a) translation of the test from English to Polish (forward translation) by two translators who were native speakers; (b) a synthesis of the two translations, which allowed discussion and elimination of discrepancies and doubts; (c) back-translation from Polish to English made by other translators than those engaged in the first stage; (d) the assessment of the translations by a team of experts and the preparation of the final version of the questionnaire. The evaluation of the translations by the team of experts was aimed at checking the quality of the translation, as well as the best adaptation of the questions to be understandable for the respondents. The last stage of the adaptation was testing the translated questionnaire during fieldwork among people who gamble regularly.
2.1.1. The cognitive interview
The quality of the translation was assessed during the cognitive interview. This is a method that allows for identification of the differences in understanding questions between researchers and respondents (Willis & Artino, 2013). The understanding of the questions was assessed by a group of randomly selected people. Answers given by these respondents were not included in the analyses. In the resulting cognitive interview, a few questions were slightly changed.
After the end of the study, a short questionnaire was sent to interviewers. They were asked about any difficulties during the screening test, problems with understanding of the questions by respondents and their subjective feelings.
2.2. The process of validation of screening tests
2.2.1. Recruitment and training interviewers
Tests were conducted by interviewers (n = 13) who were also addiction therapists, treating gambling disorders. We assumed that therapists, once they have completed certification training, use a standardized methodology for making diagnoses. Methods used by therapists for making diagnoses were discussed during training conducted by a psychiatrist, which allowed for standardization of the process of diagnosis. As a result, interviewers, when asked a series of questions about gambling, could mark the result of the diagnosis in the report of the study.
Due to the lack of adapted and validated screening tools for gambling disorders (lack of a so-called ‘gold standard’), this was a way that allowed reference of the result obtained in the screening test to the external value. In addition to the training that allowed for the standardization of the method of making a diagnosis, an instruction containing a protocol of the study was prepared. The training was recorded and circulated to all interviewers/therapists.
2.2.2. The place of conducting of the studies and sample selection
The studies were conducted in Poland, in Masovian Province (Warsaw and surroundings), Kuyavian-Pomeranian Province (Toruń and surroundings) and Świetokrzyskie Province (Starachowice and surroundings), during the second half of 2016. The inclusion criteria for respondents were their age of majority, the ability to give conscious consent to participation in the study and gambling at least once in the three months before the study.
There were four possible methods of reaching respondents, due to potential problems with recruitment. Interviewers were told to use these methods in an established order, as follows. The recommended method was to conduct the study in gambling venues; interviewers/therapists recruited respondents in legal venues, like casinos, slot machines arcades, betting points, horse races and so on. When interviewers faced difficulties reaching respondents, they were told to post advertisements on social media. The next way of reaching respondents was to use a snowball method. People who gambled regularly were recruited as a result of other respondents’ information or the interviewer’s knowledge that this particular person met inclusion criteria. The final way of finding respondents was to recruit people in gambling disorder treatment facilities.
Most of the respondents – 38.3% (n = 113): 23.5% (n = 8) women and 40.4% (n = 105) men – were recruited in addiction treatment facilities. About 30% of respondents – 29.8% (n = 88): 17.6% (n = 6) women and 31.2% (n = 81) men – were recruited in gambling venues. About a quarter of study participants – 26.8% (n = 79): 47.1% (n = 16) women and 24.2% (n = 63) men – were recruited with the use of the snowball method. Only 2% of respondents – 8.8% (n = 3) women and 1.2% (n = 3) men – were reached via social media (for example Facebook, Twitter)
2.3. Statistical analysis
Statistical analysis was carried out using a SPSS program. Following this, the validity and reliability were tested. The factor validity was assessed with exploratory (EFA) and confirmatory factor analysis (CFA): EFA was used to examine the structure, and CFA to verify a single-factor structure of the Polish version of the scale. For this purpose data from the entire sample (n = 300) were randomly divided into two subgroups (n = 150 each): EFA was performed in one group and CFA in the other.
The construct validity of the tool, both convergent and discriminant, was evaluated by the analysis of correlation coefficients between its results and the results of selected measurement tools. To assess the predictive power of the tests, the receiver operating characteristic (ROC) curve was used.
The sensitivity and specificity of the screening tests were calculated. Sensitivity measures the probability that the person gambling problematically will be identified on the tests. Specificity is defined as the probability that someone who gambles occasionally will be identified correctly. On the basis of the analysis of sensitivity and specificity, the critical value of scores from the screening tests was established. Above it, the result of the screening was considered positive. Positive predictive value (PPV) and negative predictive value (NPV) were also calculated.
The PPV is the percentage of people who have a problem, in this case a gambling disorder, among the cases indicated by the screening test as having a problem, while the NPV is understood as the percentage of people who have no problem among people indicated by the screening test as those who do not have a problem.
2.4. The ethical aspect of the study
The Bioethical Commission of the Institute of Psychiatry and Neurology in Warsaw, Poland approved the study. Participation in the study was voluntary; respondents were not paid. All respondents were given detailed information about the study, its aims, sources of financing, the confidentiality of data, and benefits and risks related to participation in the study. People who gamble are a group that is difficult to reach (especially people outside the health care), so researchers resigned of signing consent form by respondents towards statement of the interviewers about providing detailed information about the study and receiving oral consent by respondents on participation. Additionally, every respondent had ensured confidentiality by keeping their screening tests in closed envelopes.
3. Results
3.1. Characteristics of the sample and the places of recruitment of respondents
In total, 300 respondents were examined in this study: mainly men (88.3%), and only nearly 12% women. The average age of an examined person was 34.46 years (SD = 11.073). The largest group (37.7%) were 25–34 years old; and the smallest group (1.7%) were the oldest participants, 65 years old and older. Respondents mainly had a secondary education (37.1%); one in three (29.1%) had a university education, one in five (19.7%) a vocational qualification, and one in seven (14.1%) a primary or lower secondary education. Slightly more than half of respondents were in relationships (formal: 31.8%; informal: 21.7%), one in three (34.8%) were single, and one in 10 (10.7%) were divorced. Most were employed; only one in 10 (11.8%) were unemployed. More than half of respondents lived in large cities with more than 100,000 residents (58%), and only one in 20 (3.7%) lived in small towns of no more than 1000 residents (Table 1).
Table 1.
Sociodemographic characteristic of the respondents.
| Categories | Respondents |
|
|---|---|---|
| N | % | |
| Sex | ||
| Female | 35 | 11.7 |
| Male | 264 | 88.3 |
| Age (years; M = 34.46, SD = 11.073) | ||
| 18–24 | 55 | 18.5 |
| 25–34 | 112 | 37.7 |
| 35–44 | 85 | 28.6 |
| 45–54 | 24 | 8.1 |
| 55–64 | 16 | 5.4 |
| ≥65 | 5 | 1.7 |
| Education | ||
| Primary/lower secondary | 42 | 14.1 |
| Vocational | 59 | 19.7 |
| Secondary | 111 | 37.1 |
| University degree | 87 | 29.1 |
| Marital status | ||
| Married | 95 | 31.8 |
| In relationship | 65 | 21.7 |
| Separated | 2 | 0.7 |
| Divorced | 32 | 10.7 |
| Widower | 1 | 0.3 |
| Single | 104 | 34.8 |
| Employment | ||
| School-age student | 9 | 3.0 |
| Student | 13 | 4.3 |
| Full time worker | 138 | 45.4 |
| Part-time or casual worker | 50 | 16.4 |
| Self-employed | 34 | 11.2 |
| On disability pension | 16 | 5.3 |
| Retired | 7 | 2.3 |
| Housekeeper | 1 | 0.3 |
| Unemployed | 36 | 11.8 |
| Conurbation (no. of residents) | ||
| <1000 | 11 | 3.7 |
| 1000–10,000 | 18 | 6.1 |
| 10,000–25,000 | 9 | 3.1 |
| 25,000–50,000 | 58 | 19.7 |
| 50,000–100,000 | 28 | 9.5 |
| <100,000 | 171 | 58.0 |
3.2. The adaptation of screening tests
Most of interviewers declared that respondents had no problems with replying to the questions on the PGSI. If there was a problem, it was with the question connected to health problems resulting from gambling (Question 6: In the last 12 months how often have you felt that gambling has caused you health problems, including stress and anxiety?). Respondents did not connect those health problems with gambling. The next most confusing question was Question 7 (In the last 12 months how often have people criticized your betting or told you that you have a gambling problem, whether or not you thought it was true?). Respondents thought that the question was too complex. Sometimes they had to read it again to understand it. Problems with the interpretation of questions related to Questions 8 (In the last 12 months how often have you felt your gambling has caused financial problems for you or your household?) and 9 (In the last 12 months how often have you felt guilty about the way you gamble or what happens when you gamble?). In the first case, there was a problem with understanding the term financial problems; respondents did not connect it with borrowing money from their parents, but with big debts, or being completely broke. In the second case, they had a problem with interpreting the term ‘feeling guilty’ about their way of gambling or situations resulting from gambling.
Respondents did not have any problems with understanding the Lie/Bet questionnaire. No one reported any doubts about it.
3.3. Psychometric properties of the screening tests
3.3.1. Problem Gambling Severity Index (PGSI)
The results of the screening test show that more than a half (51%, n = 153) of respondents were people who might have a problem gambling. Fewer than 10% (n = 27) had no problem at all; more than 12% (n = 37) had a light severity, and almost 30% (n = 83) had a medium severity of gambling disorders.
Before beginning the factor analysis, the adequacy of the data was tested with the Kaiser–Meyer–Olkin (KMO) test. The results of the KMO test (KMO = .93) and of Bartlett’s sphericity test, χ2(8) = 1538.7, p < .001, gave information about sufficient sampling adequacy. Both the scree plot and Kaiser criterion indicated only one factor that explains 84% of the scoring variance, as also noted in the original version. All the items were found to have factor loadings above .6. The confirmatory analysis of the assumed one-factor model (without correlation of measurement errors) also achieved a satisfactory goodness of fit, χ2(27) = 138.31, p = .001 (comparative fit index, CFI = .89; Tucker–Lewis Index, TLI = .81; root mean square error of approximation, RMSEA = .36, 90% confidence interval, CI [.02, .14]; standardized root mean square residual, SRMR = .01). The results of the EFA and CFA are given in Table 2.
Table 2.
Factor loadings of the PGSI items obtained in the exploratory and confirmatory factor analyses.
| Items | Factor loadings |
|
|---|---|---|
| EFA | CFA | |
| 1. In the last 12 months how often have you bet more than you could really afford to lose? | .60 | .65 |
| 2. In the last 12 months how often have you needed to gamble with larger amounts of money to get the same feeling of excitement? | .54 | .60 |
| 3. In the last 12 months how often have you gone back another day to try and win back the money you lost? | .63 | .71 |
| 4. In the last 12 months how often have you borrowed money or sold anything to get money to gamble? | .59 | .61 |
| 5. In the last 12 months how often have you felt that you might have a problem with gambling? | .60 | .65 |
| 6. In the last 12 months how often have you felt that gambling has caused you health problems, including stress and anxiety? | .60 | .63 |
| 7. In the last 12 months how often have people criticized your betting or told you that you have a gambling problem, whether or not you thought it was true? | .51 | .59 |
| 8. In the last 12 months how often have you felt your gambling has caused financial problems for you or your household? | .71 | .77 |
| 9. In the last 12 months how often have you felt guilty about the way you gamble or what happens when you gamble? | .67 | .71 |
Note: PGSI = Problem Gambling Severity Index; EFA = exploratory factor analysis; CFA = confirmatory factor analysis.
In the PGSI test, Cronbach’s coefficient was α = .837. The values of the α coefficients that would be obtained after eliminating particular questions of the test did not indicate the necessity of removing any of them. Thus, the internal consistency of the test can be considered satisfactory. The ROC curve suggests very high predictive power of the test. The area under the curve is .969 (95% CI [.951, .987]; Figure 1, Table 3).
Figure 1.
Receiver-operating characteristic (ROC) curve of the Problem Gambling Severity Index (PGSI).
Table 3.
Predictive power of the PGSI.
| Area under ROC curve | SE | Asymptotic p | Asymptotic 95% CI |
|
|---|---|---|---|---|
| Lower bound | Upper bound | |||
| .969 | .009 | .000 | .951 | .987 |
Note: PGSI = Problem Gambling Severity Index; ROC = receiver-operating characteristic; CI = confidence interval.
Values for the sensitivity and specificity of the PGSI test are located, respectively, on levels .914 and .910 (Table 4). The combination of these indicators for cumulative distribution of test points indicates 7 points as the optimal cutoff point. It means that getting 7 points and more in the screening test indicates the occurrence of problem gambling. The PPV of the test was .925, and the NPV was .896.
Table 4.
Sensitivity and specificity of the PGSI.
| Result of the screening test | Sensitivity | Specificity |
|---|---|---|
| ≤1 | 1.000 | .203 |
| ≤2 | 1.000 | .323 |
| ≤3 | .994 | .474 |
| ≤4 | .994 | .632 |
| ≤5 | .969 | .729 |
| ≤6 | .951 | .857 |
| ≤7 | .914 | .910 |
| ≤8 | .895 | .947 |
| ≤9 | .846 | .970 |
| ≤10 | .722 | .977 |
| ≤11 | .648 | .985 |
| ≤12 | .568 | .985 |
| ≤13 | .488 | .992 |
| ≤14 | .395 | .992 |
| ≤15 | .302 | 1.000 |
| ≤16 | .235 | 1.000 |
| ≤17 | .179 | 1.000 |
| ≤18 | .154 | 1.000 |
| ≤19 | .117 | 1.000 |
| ≤20 | .086 | 1.000 |
| ≤21 | .068 | 1.000 |
| ≤22 | .049 | 1.000 |
| ≤23 | .037 | 1.000 |
| ≤24 | .025 | 1.000 |
| ≤25 | .019 | 1.000 |
| ≤26 | .012 | 1.000 |
Note: PGSI = Problem Gambling Severity Index.
Shaded value represents threshold for occurring of Gamblin problems.
3.3.2. Lie/Bet Questionnaire
The Lie/Bet test has only two questions. Results of the questionnaire showed that three quarters (75%, n = 225) of respondents qualified as needing further, deeper specialist diagnosis (they answered ‘yes’ to at least one question).
The results of the KMO test (KMO = .51) and the Bartlett’s sphericity test, χ2(1) = 59.7, p < .001, gave information about sufficient sampling adequacy. Both the scree plot and Kaiser criterion indicated only one factor that explains 71.5% of the scoring variance, as also noted in the original version. All the items were found to have factor loadings above .8. The results of the EFA are given in Table 5.
Table 5.
Factor loadings of the Lie/Bet items obtained in the exploratory factor analysis.
| Items | Factor loadings EFA |
|---|---|
| 1. Have you ever had to lie to people important to you about how much you gambled? | .85 |
| 2. Have you ever felt the need to bet more and more money? | .86 |
Note: EFA = exploratory factor analysis.
The Cronbach coefficient was α = .602. The ROC curve shows a high predictive power of the test. The area under the curve is .907 and is close to 1 (95% CI [.870, .943]; Figure 2, Table 6).
Figure 2.
Receiver-operating characteristic (ROC) curve of the Lie/Bet Questionnaire.
Table 6.
Predictive power of Lie/Bet Questionnaire.
| Area under ROC curve | SE | Asymptotic p | Asymptotic 95% CI |
|
|---|---|---|---|---|
| Lower bound | Upper bound | |||
| .907 | .018 | .000 | .870 | .943 |
Note: ROC = receiver-operating characteristic; CI = confidence interval.
Combination of the values for sensitivity (.815) and specificity (.932) of the Lie/Bet Questionnaire (see Table 7) for cumulative distribution points indicates 2 as the test’s optimal cutoff value. It means that getting 2 points on the test advises respondents of the need for an in-depth diagnosis by a specialist. The PPV of the test was .936, and NPV was .805.
Table 7.
Sensitivity and specificity of Lie/Bet Questionnaire.
| Result of the screening test | Sensitivity | Specificity |
|---|---|---|
| ≤1 | .969 | .511 |
| ≤2 | .815 | .932 |
Shaded value represents threshold for occurring of Gamblin problems.
4. Discussion
Screening tests are conducted on a general population or on a high-risk group of people. Preliminary diagnosis is possible by using screening tests, but requires confirmation using more detailed methods of diagnosis. Results of studies show that there is a necessity of using screening tests by different professional groups – for example, by general practitioners and social workers (Sullivan et al., 2000; Wieczorek et al., 2017). For general practitioners, tests may be a tool to find people at risk of gambling disorders among their patients. In Poland no scale assessing problem gambling has been validated to date. Those that are used have not been adapted and validated for this region, and therefore any results obtained with their use may be biased. Additionally, there is no screening procedure for gambling disorders that would be routinely used by general practitioners or other specialties, including psychiatrists. Moreover, such screening is not conducted in addiction treatment centers, and the results of the research show that among alcohol- and drug-dependent people the prevalence of comorbid gambling disorders is high (i.e. Pinderup, 2018).
The aim of this study was to adapt the PGSI, a commonly used international instrument, and the Lie/Bet short test for quick screening, and assess their psychometric properties when translated for use in the Polish population.
The PGSI is one of the most popular gambling disorder screening scales. It has been adapted and validated in various regions of the world. On the other hand, the Lie/Bet questionnaire is less popular but is a short scale with only two questions. Both scales have good psychometric properties in screening of gambling disorders.
The strength of this study is the careful consideration given to the translation and back translation of the two scales. The results of the adaptation of screening tests showed that the translations were easy to understand. There were occasional situations where some of the questions, answers or terms were unclear and needed further explanation and/or re-reading of the question. With regard to the Lie/Bet Questionnaire, respondents had no comments.
The results of validation of screening tests showed differences between interpretations of the results in comparison to manuals (instructions) in original versions. According to the validation of the PGSI, to establish an initial diagnosis for the identification of people who are in the risk group for meeting criteria for gambling disorders, 7 points are needed. But according to the original version it is 8 points. The manual of the Lie/Bet Questionnaire recommends deepening the diagnosis by a specialist when at least one answer is positive. Results of our study show that deepening the diagnosis is needed only when both of the answers are positive. This needs confirmation by further studies where the psychiatrist will be engaged as a person who may conduct a diagnosis. Once tests have been adapted and validated, the next step should be their dissemination to organizations and institutions in the health care and welfare area, as well as in institutions responsible for providing help for high-risk populations where gambling disorders may appear. Institutions should encourage, for example, general practitioners, therapists and social workers to conduct the tests and implement them as a tool that helps recognize a patient’s situation.
Comparison of parameters argues in favor of the PGSI. The measurement with this test is characterized by the highest sensitivity and specificity parameters, which are also balanced best. Beyond that, the PGSI is short and easy to administer, and its fulfillment does not require a great deal of time. The Lie/Bet Questionnaire has worse psychometrical parameters, but it is shorter and easier to use. It is worth noting that although most of the parameters are worse for the Lie/Bet Questionnaire, the differences between the tests are minor. Using the Lie/Bet Questionnaire is worth considering when implementation of more advanced tests is too difficult and time consuming, for example during a standard visit to a general practitioner.
The study has its limitations. One of these may be the sample size, which may seem too low. In validation studies, in which there is no ‘gold standard’ which allows verification of results obtained using a validated tool, the only option is a diagnosis by a specialist – a psychiatrist or addiction therapist. Unfortunately, this is a very expensive procedure, and the project budget allowed only for conducting 300 tests, the results of which were verified by an addiction therapist. On the other hand, a similar size of the sample was used by Stochel et al. (2015). There were also validations that were made with much smaller samples (Klecka et al., 2017). In addition, adaptation and validation of the PGSI screening test conducted in Italy on a large sample of 5300 individuals did not verify the results obtained using other tools or specialist diagnosis (Colasante et al., 2013).
Another limitation of the study may be the small number of women in the sample. A population study conducted in 2015 showed that fewer women gamble than men (45.9% vs. 54.1%, respectively; CBOS Report, 2015). Thus, such an imbalance of gender in our study does not allow us to compare our sample with the general population and the sample of gamblers in Poland.
Several benefits result from the study. The psychometric properties of two screening tests were assessed. PGSI and Lie/Bet tests are short and easy to apply, so they can be implemented in various types of institution: in primary health care facilities to conduct screening among patients, in alcohol- and drug-dependence treatment facilities to recognize comorbid gambling disorders, and in social welfare centers to be used by social workers among their clients.
Ethical standards
Declaration of conflicts of interest
Łukasz Wieczorek has declared no conflicts of interest.
Daria Biechowska has declared no conflicts of interest.
Katarzyna Dąbrowska has declared no conflicts of interest.
Janusz Sierosławski has declared no conflicts of interest.
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Ethical approval to conduct the study was obtained from Bioethical Commission of the Institute of Psychiatry and Neurology, Warsaw, Poland (ref. 14/2016).
Informed consent
Informed consent was obtained from all individual participants included in the study
Funding Statement
This work was supported by Found of Solving of Gambling Problems being in disposal of the Ministry of Health, grant: 3/HM/2016.
This work was supported by the Found of Solving of Gambling Problems of the Ministry of Health [grant number 3/HM/2016].
References
- American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). [Google Scholar]
- American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed.), Text Revision (DSM-IV-TR). [Google Scholar]
- American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). [Google Scholar]
- Beaton, D. E., Bombardier, C., Guillemin, F., & Ferraz, M. B. (2000). Guidelines for the process of cross-cultural adaptation of self-report measures. Spine, 25(24), 3186–3191. 10.1097/00007632-200012150-00014 [DOI] [PubMed] [Google Scholar]
- CBOS Report (2012). Estimation of the prevalence and identification of risk factors and protective factors of gambling, including problem (pathological) gambling and other behavioral addictions [Oszacowanie rozpowszechnienia oraz identyfikacja czynników ryzyka i czynników chroniących w odniesieniu do hazardu, w tym hazardu problemowego (patologicznego) oraz innych uzależnień behawioralnych]. Public Opinion Research Centre [Centrum Badania Opinii Społecznej CBOS]. [Google Scholar]
- CBOS Report (2015). Estimation of the prevalence of selected behavioural addictions and correlation analysis between the occurrence of behavioural addictions and the use of psychoactive substances [Oszacowanie rozpowszechnienia wybranych uzależnień behawioralnych oraz analiza korelacji pomiędzy występowaniem uzależnień behawioralnych a używaniem substancji psychoaktywnych]. Centrum Badania Opinii Społecznej [Public Opinion Research Centre] CBOS. [Google Scholar]
- Colasante, E., Gori, M., Bastiani, L., Siciliano, V., Giordani, P., Grassi, M., & Molinaro, S. (2013). An assessment of the psychometric properties of Italian version of CPGI. Journal of Gambling Studies, 29(4), 765–774. 10.1007/s10899-012-9331-z [DOI] [PubMed] [Google Scholar]
- Evans, L., & Delfabbro, P. H. (2005). Motivators for change and barriers to help-seeking in Australian problem gamblers. Journal of Gambling Studies, 21(2), 133–155. 10.1007/s10899-005-3029-4 [DOI] [PubMed] [Google Scholar]
- Ferris, J., & Wynne, H. (2001). The Canadian Problem Gambling Index user manual. Report to the Canadian inter-provincial task force on problem gambling. The Canadian Centre on Substance Abuse. [Google Scholar]
- Fortune, E. E., & Goodie, A. S. (2010). Comparing the Utility of a Modified Diagnostic Interview for Gambling Severity (DIGS) with the South Oaks Gambling Screen (SOGS) as a Research Screen in College Students. Journal of Gambling Studies/Co-Sponsored by the National Council on Problem Gambling and Institute for the Study of Gambling and Commercial Gaming, 26(4), 639–644. 10.1007/s10899-010-9189-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gambling disorders and other so-called behavioral addictions [Zaburzenia uprawiania hazardu i inne tak zwane nałogi behawioralne ] (n.d). Retrieved October 2, 2019, from, https://www.goodreads.com/work/best_book/63114561-zaburzenia-uprawiania-hazardu-i-inne-tak-zwane-na-ogi-behawioralne
- Götestam, K. G., Johansson, A., Wenzel, H. G., & Simonsen, I.-E. (2004). Validation of the Lie/Bet screen for pathological gambling on two normal population data sets. Psychological Reports, 95(3 Pt 1), 1009–1013. 10.2466/pr0.95.3.1009-1013 [DOI] [PubMed] [Google Scholar]
- Griffiths, M. D., & Nazari, N. (2020). Psychometric validation of the Persian version of the problem gambling severity index. Int J Ment Health Addiction. 10.1007/s11469-020-00336-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- ICD 11 webpage available on 03.08 (2020). https://icd.who.int/browse11/l-m/en#/http://id.who.int/icd/entity/1041487064
- Johnson, E. E., Hamer, R., Nora, R. M., Tan, B., Eisenstein, N., & Engelhart, C. (1997). The Lie/Bet Questionnaire for Screening Pathological Gamblers. Psychological Reports, 80(1), 83–88. 10.2466/pr0.1997.80.1.83 [DOI] [PubMed] [Google Scholar]
- Klecka, M., Janas-Kozik, M., Jelonek, I., Siwiec, A., & Rybakowski, J. (2017). Validation of the polish version of the Washington 4-digit diagnostic code for the assessment of fetal alcohol spectrum disorders. Psychiatria Polska, 51(2), 335–347. 10.12740/PP/OnlineFirst/62014 [DOI] [PubMed] [Google Scholar]
- Lopez-Gonzalez, H., Estévez, A., & Griffiths, M. D. (2018). Spanish validation of the Problem Gambling Severity Index: A confirmatory factor analysis with sports bettors. Journal of Behavioral Addictions, 7(3), 814–820. 10.1556/2006.7.2018.84 [DOI] [PMC free article] [PubMed] [Google Scholar]
- National Health Fund (2016). Data on number of patients and costs of treatment. Available on 30th November 2016. [Google Scholar]
- Petry, N. M. (2005). Pathological gambling: Etiology, comorbidity, and treatment. American Psychological Association. [Google Scholar]
- Pinderup, P. (2018). Challenges in working with patients with dual diagnosis. Advances in Dual Diagnosis, 11(2), 60–75. 10.1108/ADD-11-2017-0021 [DOI] [Google Scholar]
- Potenza, M. N., Fiellin, D. A., Heninger, G. R., Rounsaville, B. J., & Mazure, C. M. (2002). Gambling: An addictive behavior with health and primary care implications. Journal of General Internal Medicine, 17(9), 721–732. 10.1046/j.1525-1497.2002.10812.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- So, R., Matsushita, S., Kishimoto, S., & Furukawa, T. A. (2019). Development and validation of the Japanese version of the problem gambling severity index. Addictive Behaviors, 98, 105987. 10.1016/j.addbeh.2019.05.011 [DOI] [PubMed] [Google Scholar]
- Stochel, M., Janas-Kozik, M., Zejda, J., Hyrnik, J., Jelonek, I., & Siwiec, A. (2015). [Validation of ORTO-15 Questionnaire in the group of urban youth aged 15-21]. Psychiatria Polska, 49(1), 119–134. 10.12740/PP/25962 [DOI] [PubMed] [Google Scholar]
- Sullivan, S., Arroll, B., Coster, G., Abbott, M., & Adams, P. (2000). Problem gamblers: Do GPs want to intervene? The New Zealand Medical Journal, 113(1111), 204–207. [PubMed] [Google Scholar]
- Toneatto, T., & Millar, G. (2004). Assessing and treating problem gambling: Empirical status and promising trends. Canadian Journal of Psychiatry. Revue Canadienne de Psychiatrie, 49(8), 517–525. 10.1177/070674370404900803 [DOI] [PubMed] [Google Scholar]
- Wieczorek, Ł., Dąbrowska, K., & Moskalewicz, J. (2017). Recommendations for improving the availability and the quality of treatment service for gambling disorder persons. Alcoholism and Drug Addiction, 30(2), 113–136. 10.5114/ain.2017.70289 [DOI] [Google Scholar]
- Willis, G., & Artino, A. (2013). What do our respondents think we’re asking? Using cognitive interviewing to improve medical education surveys. Journal of Graduate Medical Education, 5(3), 353–356. 10.4300/JGME-D-13-00154.1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- World Health Organization. (2004). ICD-10 : international statistical classification of diseases and related health problems (10th revision, 2nd ed.). [Google Scholar]
- World Health Organization. (2018). International classification of diseases for mortality and morbidity statistics (11th Revision). Retrieved from https://icd.who.int/browse11/l-m/en [Google Scholar]