In the following section the results are highlighted in detail and related to the variables used in this research. These are variables drawn from the “Theory of Reasoned Action”. These variables are hypothesized as being predictive of the Intention to sign-up for the non-anonymous variant. Most variables were measured on a 5-point Likert scale from “strongly disagree” to “strongly agree” with a mean rate of 3. Mean scores smaller than three are entitled as low and scores higher than three as high.
4.2.1 Perceived severity
One of the variables which is believed to influence the Intention to sign-up for the non-anonymous variant is Perceived Severity of alcohol abuse. Perceived Severity measured the level of seriousness and urgency of the client to undergo the online treatment. It had a mean of 4.41. This was exemplified in the response on one of the questions measuring this dimension: “Ik denk dat ik meer drink dan de meeste mensen.” The mean response to this question was 4.34 which is very high.
Perceived Self-Efficacy is another variable which was hypnotized to exert some influence on the intention to sign-up for the non-anonymous treatment variant. It measured the person’s belief in his or her ability to succeed in the aspired behavior which, in this case, is changing the drinking pattern in this case. The item measuring this variable was “Ik denk dat ik goed in staat ben mijzelf ertoe te zetten mijn huidige alcoholgebruik te veranderen” and had a mean score of 3.74
Subjective Norm as a variable was derived from the Theory of Reasoned Action and is a predictor of Intention to sign up for the non-anonymous treatment. It measured the normative influence or the influence of significant others on participants’ decisions with regard to choosing to undergo the treatment or not. The mean score was 2.7. This tells us that most people did not agree with the statement “De meeste mensen zouden graag willen dat ik deze behandeling volg”. This is true for 45.2% who do “not agree” or “strongly disagree” with this point.
Motivation to change measured the degree to which the participants are willing to change their behavior in general. The item to measure this variable was “Ik ben heel gemotiveerd om mijn drinkgedrag te veranderen.” The mean score for this item was 4.05 which is a relatively high score with 70.9% of the people agreeing to being highly motivated to change their drinking pattern.
Outcome Beliefs were also derived from the Theory of Reasoned Action. It is theorized as being a predictor of Intention to sign-up for the non-anonymous treatment and had a mean score of 3.95. This variable measured the impact of several outcome beliefs people were holding about the purpose of the online treatment. The goal of this construct was also to find out which Outcome Beliefs were of greatest and least importance for participants. The items with the highest means and therefore most important for people were about wanting to learn techniques to deal with the problem “Ik wil technieken leren, die mij helpen om met mijn alcoholprobleem om te gaan” (4.55), aiming to regain control over oneself “Ik wild at de behandeling mij help tom weer controle te krijgen over mijn drinkgedrag” (4.53), and wanting to lower the general alcohol consumption “Ik wil minder alcohol drinken” (4.50). The least important outcome belief was to improve the relationship with family and friends “Ik wil de relatie met mijn vrienden en familie verbeteren” (3.06). Two other items also scored fairly low for wishing to stop the alcohol consumption “Ik wil stoppen met het drinken van alcohol” (3.08) and wanting to work more efficiently at one’s job “Ik wil beter kunnen functioneren op mijn werk” (3.13).
Barriers towards signing up for the non-anonymous variant
Another variable in the modified model is Barriers against signing up for the non-anonymous variant. It was hypothesized that this variable would influence the intention to sign up for the non-anonymous variant. This construct examined the barriers derivable from taking all the probable negative impairments when signing up for the non-anonymous variant. The mean score for this variable is 3.19.
The item with the highest mean was “Ik kan kiezen dus dan liever anonym” (4.42) which states that the opportunity to choose between anonymous and non-anonymous is a great barrier when it comes to feeling intended to sign-up for the non-anonymous variant. There were two other items with relative high means. These items relate to the problem of the employer “Ik wil niet dat mijn werkgever weet dat ik alcohol problemen heb” (4.14) and the social environment knowing about the alcohol abuse problem “Ik wil niet dat mensen uit mijn omgeving van min alcoholproblem weten” (4.13). Items with the lowest mean scores were items about having no health insurance in the Netherlands “Ik ben Nederlander maar niet in Nederland verzekerd en dan wordt de behandeling niet verged” (2.48), not having the Dutch nationality therefore not eligible to receive reimbursement from the health insurance “Ik ben geen Nederlander en anders wordt de behandeling niet verged” (2.66), not having known about the possibility to choose for another treatment option “Ik weet niet zeker of de behandeling helemaal door mijn zorgverzekering wordt vergoed” (2.71), and not be willing to pay the contribution for health insurance “Ik ben niet bereid om de eigen bijdrage te betalen” (2.76).
Information Status is one of the new variables we introduced into the model. It has been hypothesized that if participants only have insufficient information about the alternative treatment variants, that they won’t sign-up for these, because they don’t know about them or they don’t know about the implications and advantages of these alternatives. The question was whether or not participants had read the information about the three alternative treatment options, employer-paid, insurance-paid and self-paid option. The outcome percentage for not reading the information was 58.1% for the employer-paid option, 48.4% for the insurance-paid one and 51.6% for the self-payers. Also, 61.3% of the participants didn’t know that when choosing one of the alternative options, they would be able to start the treatment directly.
Intention is another variable in the Theory of Reasoned action, which we adopted for completeness. This variable measures the level of effort one is willing to exert to attain a behavioral goal. The item used to measure the Intention was “Nu u weet wat er met uw gegevens gebeurt: zou u bereid zijn om u in te schrijven voor de zorgverzekeringsvariant in plaats van anonym te blijven?” The mean score is of 2.15 the lowest score among all the variables and implies that most participants are not willing to sign up for the non-anonymous treatment even if they know a bit more about how their personal data is used and stored. 67.7% respond that they do not intend to sign up for the non-anonymous treatment in the foreseeable future and only 20.9% state that they are willing or perhaps willing to sign-up for the non-anonymous treatment anytime soon.
Solution Suggestions is another construct we believed would influence the intention to sign up for the non-anonymous treatment option. It measures the importance of different problem solving possibilities to participants seeming to prefer the anonymous variant. These participants have to wait a long time for treatment while there is the possibility of non-anonymous treatment were they could start directly. The mean score of this construct is 3.23 which shows a general tendency to the usefulness of these Solution Suggestions as the mean is higher than 3. Our goal with these items also was to discover the most influential solution suggestions for participants. The item with the highest mean contains the option of deleting all the personal details of a client after finishing the therapy “Ik zou eerder geneigd zijn om me voor de zorgverzekeringsvariant op te geven als mijn gegevens achteraf vernietigd worden” (3.87). Close to this score were the items about having more control about what happens to the personal details afterwards “Ik zou eerder geneigd zijn om me voor de zorgverzekeringsvariant op te geven als ik meer controle zou hebben over wat achteraf met mijn gegevens gebeurt” (3.73) and the possibility of not getting the general practitioner involved in the procedure “Ik zou eerder geneigd zijn om me voor de zorgverzekeringsvariant op te geven als de huisarts het niet te weten zou komen” (3.39). The item with the lowest mean stated that the enrolment for the non-anonymous variant should be more easy “Ik zou eerder geneigd zijn om me voor de zorgverzekeringsvariant op te geven als de aanmeldingsprocedure voor de zorgverzekeringsvariant makkelijke zou zijn” (2.73).
A correlation matrix was constructed showing all important constructs against each other to see whether there are any correlations between them. The demographic variables are not included because there were no significant results found. After that a regression analysis was conducted.
Correlations between psychological variables are shown in Table 4.2
In this section we only looked at linear relationships. These only show which variables are related and the direction of the relationship. However, for us to understand which independent variables determined a particular dependent variable we need to look for relationships based on causality. Causality is essential to differentiating which variables are related by chance and those that are predictive. In the next section we use multiple regression analysis to enable us to highlight causal relationships among the variables under study.
4.3.2 Regression analysis
In this section regression analysis is used to plot the significant determinant pathways between the various variables and their relationship with Intention to sign-up for the non-anonymous variant. The variables that correlated significantly with in bivariate analyses were included in the regression analysis.
In total, two variables had a significant correlation with Intention (see correlation matrix. These were Solution Suggestions (r= .307, p<.05) and Information Status (r=-.340, p<.05).
Using the regression analysis we found two main predictors of Intention to sign up for the non-anonymous treatment variant. These were, Information Status (b=-.289, p<.05) and Solution Suggestions (b=.247, p<.05). The two variables together account for about 40% (r²=.417) of the variance of Intention.
The beta figures show what each variable contributed to the model. Solution Suggestions had a positive relationship with Intention. Information Status has a negative relationship with Intention. If people indicate that they are well informed about the different variants, they are less likely for having the intention to sign up for non-anonymous variant.