Applying the theory of planned behavior to predict online addiction treatment intention

Abstract

Scant research has been published on online addiction treatment, with few studies focusing on the Theory of Planned Behavior (TPB) in relation to addiction treatment. Therefore, this study aims to explain the TPB's predictability on intention and behaviors in relation to online addiction treatment among people recovering from substance use disorder (SUD).

Methods

This descriptive-analytical study included a self-report questionnaire based on the TPB model, and was distributed to a sample of 115 people recovering from SUD, aged 18-69, 62% of whom were men.

Results

Attitude, Subjective Norms (SN), and Perceived Behavioral Control (PBC) toward online addiction treatment was significantly positive in relation to intention and past behavior of participants in online addiction treatment. Attitude and PBC were found to be significant predictors, and the TPB model was found to be significant {F (3,111) = 47.29, p < 0.01}, explaining 56% of the variance of intention for participants in online addiction treatment.

Conclusion

As online treatment is a relatively new tool in addiction treatment, professionals and treatment providers should encourage beliefs, attitudes, moral norms, and perceived behavior control to increase intentions among future participants in online addiction treatment.

Keywords:

• Theory of planned behavior
• online treatment
• telehealth
• addiction
• recovery
• Israel

Introduction

Stressors caused by COVID-19 contribute to increased substance use in the general population^1–5 and increased addictive behaviors and cravings of people who have recovered from substance use disorder (SUD).⁶ Accordantly, the demand for admission to addiction treatment services has increased.⁷ The social distancing and lockdowns caused by the pandemic as well as the fact that patients suffering from addiction are a vulnerable group during COVID-19, along with their healthcare being considered essential,⁸ has caused addiction treatment services to find more flexible ways to provide treatment in the form of turning to telehealth. As such, the use of digital communication technologies via the internet and online video platforms has emerged as a primary mode of accessing substance use-focused mutual-help group meetings.^9,10

Mental health care professionals and organizations have been relatively slow in adopting technological tools for remote psychological treatment.^11–13 In addiction treatment, telehealth was used only in about 0.1% of addiction treatment visits prior to the COVID-19 pandemic, even though about 27% of specialty addiction facilities had telehealth capabilities.¹⁴ The COVID-19 pandemic caused sudden necessitated and radical changes in mental health care delivery, as almost overnight practitioners were forced to transfer their face-to-face care practice to online channels.¹² As such, addiction treatment providers rapidly pivoted from primarily delivering in-person addiction treatment to providing telehealth treatment.¹⁵ For example, the number of online mutual-help groups supporting individuals with problematic alcohol, substance use, and other behaviors, provided by SMART Recovery Australia (SRAU), in response to the COVID-19 pandemic has increased significantly from a mere 6 pre-COVID-19 to a remarkable 132.¹⁶ The convenience of the internet and smartphones made it possible for psychiatrists and mental health centers to provide online mental health services during the COVID-19 outbreak,¹⁷ as online health tools offer a potential solution for the continuation of high-quality mental health care.^18,19

The pandemic generated an understanding that online mental services should be further developed to help maintain support of mentally ill patients during major public health events like COVID-19.²⁰ Very little has been published on the benefits of addiction treatment via telehealth. A literature review until 2021, identified only eight published studies that compared addiction treatment via telehealth with in-person treatment. Most of those studies were conducted among small samples (less than 150 patients). Seven found telehealth treatment as effective but not more effective in terms of retention and satisfaction with the treatment of substance use.¹⁴ The recent publication by SRAU highlights the positive outcomes of delivering online recovery-focused mutual-help groups, emphasizing their accessibility, acceptability, and sustainability. This sheds light on the effectiveness and advantages of using online platforms to support individuals seeking addiction recovery.¹⁶ Interview participants highlighted that telehealth reduces the time and cost to patients of participating in treatment and offers an opportunity for clinicians to observe patients’ home environment and engage with their families. On the other hand, many participants felt strongly that patients with substance use disorders need personal relationships and connectedness, which are hard to establish in virtual environments. Additionally, they noted that it is more difficult to sense how a patient is doing when meeting via telehealth and it can be challenging to keep patients focused online. Interviewees also noted that telehealth may work better for some patients and clinicians than others.¹⁴ Another recent study used national data collected before the pandemic and found that online meeting attendance was more likely among women than men as well as younger than older participants. It also found that online meeting attendance may have an appeal and be helpful to mutual-help group members who are at an earlier stage in their recovery.¹⁰

The theory of planned behavior (TPB)²¹ was utilized in the current study in an attempt to gain a theoretical understanding of online addiction treatment intention and behavior among people recovering from SUD during COVID-19. TPB theory assumes that intentions predict actual behaviors. According to TPB, intention is viewed as a function of three socio-cognitive factors: 1. Behavioral beliefs – these produce a favorable or unfavorable attitude toward the behavior; 2. Normative beliefs – these result in perceived social pressure or subjective norms (SN); and 3. Control beliefs – these give rise to perceived behavioral control (PCB). The effects of attitude as a positive or negative evaluation of the behavior and SN as the perception of whether other important individuals wish or expect the participant to behave in a certain way on intention are moderated by PCB, as the individual’s perception of how easy or difficult it is to execute the behavior. Together, these shape an individual’s behavioral intentions. The more favorable the attitude and SN, and the greater the PCB, the stronger the individual intention to perform the behavior. According to TPB theory, the three considerations of attitude, SN, and PBC should sufficiently account for all meaningful variance in behavioral intention. In the same way, other factors such as demographics, severity, and past behavior should contribute as a function of these first-order constructs and not be added to explain variance in intention.^21–23 Among the three considerations, SN was shown to be the weakest for predicting intention.^24,25 Several studies have also shown that in addition to the three socio-cognitive factors (i.e., attitude, SN, and PBC) past behavior can also have a model-independent influence on later behavior.²⁶ In addition, the latest structure recommendation to the TPB questionnaire included an examination of past behavior,^23,27 as also measured in the current study.

In fact, TPB has been supported in a variety of settings²⁸ and was found to be a useful model in explaining intentions to engage and actual engagement in several health behaviors.²² A few studies have focused on TPB in addiction settings; for example, predicting patient post-detoxification engagement in 12-step groups,²⁵ predicting condom use intention among Iranian substance users in addiction treatment centers,²⁹ and predicting substance abuse treatment completion based on the theory of planned behavior.³⁰However, little has been published on TPB and addiction treatment during the COVID-19 pandemic.

Objectives

As noted, scant research has been published on online mental health treatment in general, and online addiction therapy in particular.¹⁴ To our knowledge, this is the first study to predict by TPB participation of people recovering from SUD in online addiction treatment during the COVID-19 pandemic. This research, therefore, has the following objectives: (a) Drawing on TPB in predicting the intention to participate in online addiction treatment and (b) Testing whether intention has positive correlation with behavior. We would like to emphasize that in Israel, unlike in the United States, for example, the distinction between “addiction treatment” and “addiction recovery” is often not emphasized, especially in the terminology used. Therefore, in this paper, we will primarily use the term “addiction treatment” to encompass both treatment and recovery collectively.

Method

Participants

This study included 115 people recovering from substance use disorder. The inclusion criteria were: (a) At least 18 years of age and (b) Having participated in addiction treatment in the last year for at least three months. Participant ages ranged from 18 to 69 (Mean = 38.57, S.D. = 11.05), of whom 71 (62%) were men. Table 1 presents the demographic variables of the participants.

Table 1. Socio-demographic characteristics and type of addiction treatment (N = 115).

Variables		Total sample (N = 115)
		N	%
Age M (SD)	18-69 (range)	38.6 (11.05)
Gender	Male	71	61.7%
	Female	44	38.3%
Nationality	Jewish	94	81.7%
	Moslem	10	8.7%
	Christian	9	7.8%
	Druze	1	0.9%
	Other	1	0.9%
Marital status	Single	31	27%
	Married	45	39.1%
	In a relationship	14	12.2%
	Divorcee	20	17.4%
	Separated	5	4.3%
Employment status	Employee	85	73.9%
	Unemployment	28	24.3%
	Unpaid vacation	2	1.7%
The primary substance consumed prior to beginning the treatment	Cannabis	45	39.1%
	Cocaine	16	13.9%
	Heroin	25	21.7%
	Alcohol	13	11.3%
	Ecstasy/ MD	7	6.1%
	Prescription medicine	5	4.3%
	Other	4	3.5%
The main reason for choosing to seek treatment	Family pressure	27	23.5%
	Legal proceedings	15	13%
	Medical reason	7	6.1%
	Financial reason	7	6.1%
	My personal Decision	49	42.6%
	Other	10	8.7%
Type of addiction treatment in the last year	Individual treatment - face to face	62	53.9%
	Individual treatment – online	39	33.9%
	Group therapy in a professional therapeutic setting - face to face	45	39.1%
	Group therapy in a professional therapeutic setting – online	22	19.1%
	Self-help groups/the 12 steps - face to face	86	74.8%
	Self-help groups/ the 12 steps - online	87	75.7%

Research tools

The questionnaire included the following two sections:

A socio-demographic and participant addiction treatment questionnaire

Participants completed a socio-demographic questionnaire including age, gender, education level, marital status, religiosity, nationality, place of birth, level of income, and employment status. Additionally, the participants were asked about current addiction treatment: group or individual treatment, self-help groups or not, and online or face-to-face meetings. As previously mentioned, the distinction between “addiction treatment” and “addiction recovery” in Israel is often not emphasized. Consequently, the term “addiction treatment” encompasses both treatment and recovery as a collective concept.

TPB theoretical construct questionnaire for online addiction treatment

TPB scale variables were measured using the guidelines given by Fishbein & Ajzen ^23,27 that included 8 items under five constructs, including (a) attitude, (b) subjective norms (SN), (c) Perceived behavioral control (PBC), (d) intention, and (e) past behavior.

(a) Attitude toward online addiction treatment

Two items were designed to assess positive attitude toward online addiction treatment. Participants were asked to complete the following statement: “For me, the option to seek online treatment to treat my addiction would be”. The items were rated on a 7-point scale ranging from 1 (bad) to 7 (good) and (1 = unpleasant; 7 = pleasant). Higher scores indicated more positive attitude to online addiction treatment.

(b) Subjective norms (SN)

Two items were designed to measure SN toward online addiction treatment. SN were measured by the following statement: “Most people who are important to me think that I should participate in online addiction treatment”. The items were rated on a 7-point scale ranging from 1 (disagree) to 7 (agree) and (1 = unlikely; 7 = likely). Higher scores indicated more positive attitude to online addiction treatment.

(c) Perceived behavioral control (PBC)

Two items were designed to measure PBC. A sample item was “I am confident that I can participate in online addiction treatment”. The items were rated on a 7-point scale ranging from 1 (false) to 7 (true) and (1 = disagree; 7 = agree). Higher scores indicated more positive attitude to online addiction treatment.

(d) Intention

Intention toward online addiction treatment was assessed by using a single item. The item was “I intend to participate in online addiction treatment”. The items were rated on a 7-point scale ranging from 1 (unlikely) to 7 (likely). Higher scores indicate more intention to participate in online addiction treatment.

(e) Past Behavior

Online addiction treatment as past behavior was assessed by using a single item. The item was “In the past three months, I have participated in online addiction treatment”. The items were rated on a 7-point scale ranging from 1 (false) to 7 (true). Higher scores indicate more intention to participate in online addiction treatment.

Procedure

This study was approved by the ethical standards of the Institutional Review Board (IRB) Ariel University. Approval number AU- criminology {CRI}, 036- JUN-2021, of the first author’s institution. The questionnaire was distributed to a convenience sample of people who had recovered from SUD. We enrolled participants using the snowball sampling method and asked respondents to relay the questionnaire to friends. The questionnaire was distributed between 18 July 2021 and 10 February 2022 during COVID-19. All participants gave their informed consent and participated voluntarily in this study and were not asked to provide any identifying personal information.

Data analysis

Data analysis was conducted with SPSS ver. 25. We analyzed the data’s descriptive statistics showing the characteristics of the sample and participation in online addiction treatment. The associations between the TPB variables were examined by Pearson’s correlation. T-tests and one-way ANOVA were conducted to examine differences in TPB variables. Linear regression analysis was employed to predict behavioral intention to participate in online treatment addiction.

Results

All participants reported participation in addiction treatment for at least three months during the last year, as 81.7% reported current online addiction treatment. Among them, 19.1% reported participation in online group treatment, 33.9% in individual online addiction treatment, and 75.7% reported participation in online self-help groups such as 12 steps.

In the current study, TPB was employed to investigate participation in online addiction treatment. The participants were asked about their attitude, SN, perceived behavioral control, behavioral intention, and past behavior. The associations between the TPB variables are presented in Table 2.

Table 2. Descriptive statistics and intercorrelation between the TPB variables toward online addiction treatment (n = 115).

Variables	1	2	3	4	5	M	SD
Attitude	1	—	—	—	—	4.45	2.24
Subjective norms (SN)	**25.	1	—	—	—	3.16	1.92
Perceived behavioral control (PBC)	**62.	*18.	1	—	—	5.46	1.82
Intention	**72.	**28.	**58.	1	—	3.95	2.68
Past behavior	**65.	**25.	**56.	**75.	1	4.37	2.87

*p < 0.05 , **p < 0.01.

Table 2 shows bivariate associations between the TPB variables. Attitude (r = .72), SN (r = .28), PBC (r = .58) toward online addiction treatment were significantly positive related to Intention. In addition, attitude (r = .65), SN (r = .25), and PBC (r = .56) toward online addiction treatment were significantly positively related to past behavior toward participation in online addiction treatment.

Based on the findings, there is a negative correlation between the age variable and actual behavior (r = −0.21, p < 0.05). This indicates that as age increases, there is a decrease in the likelihood of participants seeking online addiction treatment.

Table 3 presents the gender differences in TPB variables regarding online addiction treatment.

Table 3. Gender Differences in TPB variables toward online addiction treatment (n = 115).

		male (n = 71)		female (n = 44)
Variables	t	M	SD	M	SD
Attitude	–1.69	4.19	2.41	4.87	1.89
Subjective norms (SN)	–1.37	2.97	1.96	3.47	1.85
Perceived behavioral control (PBC)	–2.15*	5.20	2.06	5.87	1.26
Intention	–1.79	3.61	2.74	4.52	2.51
Past behavior	–1.58	4.04	2.89	4.91	2.78

*p < 0.05.

Table 3 displays the significant gender differences observed in PBC regarding online addiction treatment. Specifically, women exhibited higher levels of PBC compared to men. However, no significant gender differences were identified in the other variables examined.

A t-test was conducted to examine whether there are differences in TPB variables related to online addiction treatment based on the primary substance consumed. The results indicated significant differences in PBC (t (113) = 2.05, p < 0.05). Post-hoc Bonferroni tests revealed that individuals who reported consuming alcohol had significantly higher PBC scores (M = 6.42, SD = 0.97) compared to those who consumed other substances, including cannabis, cocaine, heroin, ecstasy, prescription drugs, and others (M = 5.33, SD = 1.87).

A one-way ANOVA was performed to investigate whether there are differences in TPB variables related to online addiction treatment based on the main reason for seeking treatment. The analysis revealed a significant difference in SN (F(5,109) = 2.67, p < 0.05). Post-hoc Bonferroni tests indicated that the difference was between the “medical reason” group and the “other” group, with those citing medical problems having higher SN scores than those in the “other” category.

In order to assess the contribution of attitude, SN, and PCB to the intention to participate in online addiction treatment, a linear regression analysis was performed, as presented in Table 4.

Table 4. Linear regression analysis to predict the behavioral intention of participants in online addiction treatment (n = 115).

Variables	B	S.E.	β
Attitude	68.	10.	**57.
Subjective norms (SN)	14.	09.	10.
Perceived behavioral control (PBC)	31.	12.	**21.

**p < 0.01.

According to the results presented in Table 4, the TPB model was found to be significant {F (3,111) = 47.29, p < 0.01} and explains 56% of the variance of intention to participate in online addiction treatment. However, only two of the three variables were found to be statistically significant predictors for the intention to participate in online treatment based on the TPB model: Attitude toward online addiction treatment (p < 0.01) and PBC (p < 0.01), as SN toward online addiction treatment, was not found to be a significant predictor.

Discussion

Little is known about online addiction treatment,¹⁴ with only a few studies having examined TPB in an addiction setting.^25,29 To our knowledge, no study has yet focused on the TPB in relation to participants in addiction treatment during the COVID-19 pandemic. As such, this study fills a gap in the literature and expands our knowledge regarding online addiction treatment and TPB among people recovering from SUD during COVID-19.

This research found that attitude, SN, and PBC toward online addiction treatment were significantly positive in relation to intention and past behavior toward participants in online addiction treatment. Previous studies have shown that among the three considerations (attitude, SN and PCB), SN was shown to be the weakest for predicting intention.^24,25 In our study, linear regression analysis found that SN did not significantly predict the intention to participate in online treatment. Indeed, most of the mental and addiction treatment worldwide occurs in face-to-face meetings in various settings. Even so, research on online treatment is still scarce. However, it can be assumed that SN is the weakest for predicting intention, partly due to the newness of online addiction treatment services, with few participants and evaluation studies. Therefore, addressing SN may not be the most effective way to influence people’s intentions to participate in online addiction treatment. Professionals and treatment providers should encourage people to participate in online treatment by shifting their attitudes in a more positive way or enhance PBC by providing practical help such as online services or practice guidance as well as online trainings on how to prepare for online addiction treatment. This is important due to public perception that face-to-face health treatment is better than online health services. In contrast to SN, attitude may have a stronger predictive power for participants in online addiction treatment. This may be explained by Festinger’s ³¹ cognitive dissonance theory, which suggests that people have an inner drive to maintain their attitudes and behaviors in harmony, avoiding disharmony or dissonance. In our case, online addiction treatment was the only solution available during COVID-19. As there was no other alternative, people adjusted their attitudes toward participation in online addiction treatment. In addition, the online platform has benefits such sas availability and accessibility, especially through remote access. As such, people were given the opportunity to visually meet and stay in contact with other professionals and peer groups during periods of isolation and loneliness.

Similar to a previous study,¹⁰ also the current research indicates that as age increases, there is a decrease in the inclination to seek online treatment. This trend can be attributed to older individuals potentially having limited familiarity or comfort with technology and online platforms, which diminishes their likelihood of pursuing digital treatment options. Moreover, older individuals may exhibit a preference for more traditional or in-person methods of treatment, as they tend to prioritize face-to-face interactions and personal connections in their therapeutic journey.

In line with these findings, the current study also reveals that women displayed higher levels of PBC in relation to their participation in online addiction treatment compared to men. This observation is consistent with a previous study,¹⁰ which found that online meeting attendance was more likely among women than men. The gender differences can be attributed to the flexibility and convenience offered by online treatment. Women, who often juggle multiple responsibilities and commitments, perceive online treatment as more accessible and manageable. The flexibility allows them to engage in therapy at their own pace and from the comfort of their preferred environment. These factors contribute to women’s higher levels of PBC and their increased likelihood of participating in online addiction treatment.

As noted, the present TPB model worked well for assessing intentions and as a predictor of attending online addiction treatment among people who recovered from SUD. As demand for admission to addiction treatment services increased during COVID-19,⁷ with patients suffering from addiction being a vulnerable group and their healthcare considered essential,⁸ it is important to develop more flexible means to provide online addiction treatment. Because online treatment is a relatively new tool in addiction treatment, professionals and treatment providers should encourage beliefs, attitudes, moral norms, and perceived behaviors to increase intentions among future participants in online addiction treatment.

However, several limitations of this study should be noted. First, this study is cross-sectional in nature and relies on self-report measures. Second, people who recovered from SUD represent a small group of the population and are difficult to locate. Thus, snowball sampling can lead to a possible sampling bias. The sample is also composed of a majority of Jewish participants. Therefore, more studies are needed employing a more controlled sample with greater variety.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.