Participants were recruited using a survey link posted via emails to professional organization listserves, specifically the American Psychological Association Division 29—Society for the Advancement of Psychotherapy, the Association for Behavioral and Cognitive Therapies, and social media (ie, Twitter). The survey, which was advertised as a survey on “waiting lists and possible resources,” began with a Study Information Sheet outlining the purpose of the research, eligibility criteria, limits of confidentiality, risks and benefits, and compensation. Participants were eligible to participate if they identified as being (1) over the age of 18 years, (2) a practicing licensed mental health care professional, and (3) currently providing at least 1 hour of clinical services per week. We required participants to have to conduct at least an hour of clinic work to ensure participants were currently providing at least some clinical services, but we did not limit the study to those whose only duties were clinical work.

We received 145 survey responses. Two responses were removed after being determined to be from the same individual. Two responses were removed for not providing responses to all questions. The remaining 141 participants were included in data analysis.

Study procedures were approved by Indiana University Bloomington’s Human Subjects and Institutional Review Board (10503).

The survey, which can be found in the Open Science Framework website [32], was divided into 4 sections. Section 1 collected demographic information (eg, age, gender, race, and ethnicity) and clinical background (eg, state or country of licensure, level of education, clinical practice setting, psychotherapy theoretical orientation, satisfaction with clinical work, years of clinical experience, and average hours of clinical services provided per week).

Section 2 was designed to address aim 1 of our study. In this section, participants were asked to provide information on their current waiting list practices. This included whether their current clinical setting kept a treatment waiting list (vs not keeping a waiting list or scheduling appointments several months in advance), the estimated time patients spent on a waiting list, the effect of the COVID-19 pandemic on waiting time, and current actions taken for patients on a treatment waiting list.

In section 3, participants were provided with a brief description of the following LITs: unguided bibliotherapy, guided bibliotherapy, unguided web-based self-help, guided web-based self-help, and patient support groups. We chose these LITs because they have been relatively well researched, and research suggests they are effective. Descriptions of each LIT type were provided to ensure that respondents were equally familiar with each intervention. To address aim 2 of our study, for each LIT, respondents were asked to rate their perceived effectiveness (eg, “I believe this option would be effective for patients on a treatment waiting list”), availability (eg, “I believe this option is available and accessible to use with patients on my waiting list”), and willingness to use LIT (eg, “I am willing to use this modality for patients on a waiting list”). Responses were given on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree).

Section 4 addressed aim 3 of our study by asking respondents if they would like to receive any additional information about the following LITs: bibliotherapy, web-based self-help, or patient support groups. They were subsequently offered information on each LIT, and we tracked which participants chose to receive more information (ie, whether they engaged in information-seeking behavior).

Missingness in demographic and clinical variables was relatively low (0%-1.4%), with the exception of age (13.5%). Missingness in the attitude variables was greater (10.6%-12.1%), reflecting survey dropout (ie, participants who answered no subsequent questions). We completed regression analyses with the original data set and an imputed data set. The latter imputed missing values for all variables included in the regression analyses using a machine learning algorithm with random forests using the R package “missForest” [33].

Analyses were conducted using the R programming language (version 4.2.1; the R Core Team) [34]. To describe waiting list practices, we present response frequencies and descriptive statistics (aim 1). We compared descriptive statistics and response frequencies of providers’ perceived efficacy, availability and accessibility, as well as willingness to use the different LITs (aim 2).

To explore predictors of willingness to use digital and nondigital LITs (aim 3), we ran 5 linear regressions, one for each LIT under consideration. Willingness to use each specific intervention was regressed on demographic information (ie, gender, age, and education), professional background (ie, theoretical orientation and practice setting), and clinical variables (ie, the use of a waiting list, clinical satisfaction, and clinical hours per week). In these models, an average “willingness” response variable was included in the regression to control for participant willingness to use other interventions, excluding the one being predicted. Regressions were completed using the original data set and the imputed data set.

To explore predictors of information seeking for digital and nondigital LITs (aim 3), we conducted a series of regressions to explore demographic, professional background and attitudinal predictors of requesting additional information. Specifically, we conducted 3 binomial logistic regressions with the dependent variables of requesting additional information (ie, selecting information vs not selecting information) on the following: (1) web-based self-help, (2) bibliotherapy, and (3) patient support groups. Each model included demographic information (eg, gender, education, and age), clinical variables (eg, theoretical orientation, clinical satisfaction, years of experience, clinical hours, and presence of a waiting list), modality-specific attitudes (eg, willingness to use the specific intervention and perceived availability and accessibility of the specific intervention), and controls for overall rating tendency (eg, average willingness to use other interventions and average availability or accessibility of other interventions).

We report how we determined our sample, all data exclusions, all manipulations, and all measures in the study. All data, analysis code, and research materials are available on the Open Science Framework website [32]. This study’s design and its analysis were not preregistered. No other papers currently use these data.

Demographic and clinical variables of the sample are summarized in Table 1. The sample primarily included female, non-Hispanic, White participants. Participants’ average age was about 39 years. Most participants in the sample had a PhD degree and were prescribed to an orientation (eg, cognitive or third-wave behavioral therapy) related to cognitive behavioral therapy (CBT). About half of the participants were employed in a private practice setting with an average of 10 (SD=10.4) years of clinical experience.

The majority of survey respondents (n=94, 69.1%) endorsed keeping a formal treatment waiting list (Table 2) with the estimated waiting time averaging about 13 weeks. Others reported not keeping a waiting list but scheduling patients “a couple of months in advance” (n=23, 16.9%). Most of the respondents who endorsed keeping a waiting list also noted their estimated waiting time had been impacted by the COVID-19 pandemic (n=68, 73%) and estimated waiting times around 9 weeks prior to the pandemic.

The majority of participants who kept a waiting list or scheduled patients for months in advance endorsed taking names or contact information, completing an unstructured brief assessment, and providing referrals. Few reported completing a structured assessment or provided brief psychoeducation, and even fewer (n=13-20, <20%) reported providing information on apps, books, or support groups.

Figure 1 shows respondents’ ratings of willingness to use, perceived efficacy, and availability of different interventions. Average responses were generally between 3 (“neutral”) and 4 (“somewhat agree”). The exception to this was the perceived low accessibility and availability of patient support groups (mean 2.9, SD 1.2). Average attitudes were highly correlated; specifically, the willingness to use LITs was related to its perceived efficacy (r₁₂₂= 0.76, 95% CI 0.67-0.83; P<.001).

We used multiple linear regressions to identify demographic, clinical, waiting list, and attitude variables predicting participants’ ratings of willingness to use a LIT (Table S1 in Multimedia Appendix 1). Across models, averaged willingness to use other LITs predicted the specific likelihood of willingness to use another LIT. For example, the willingness to use unguided bibliotherapy was strongly predicted by the average willingness to use support groups, guided bibliotherapy, and web-based self-help (β=.65, 95% CI 0.45-0.84).

Other predictors of willingness to use a LIT varied across the LITs probed. Compared to individuals in other practice settings, individuals in private practice were more willing to use unguided bibliotherapy (β=.22, 95% CI 0.02-0.42) and less willing to use guided bibliotherapy (β=–.20, 95% CI –0.35 to –0.04). Additionally, individuals who reported a CBT theoretical orientation (vs not reporting one) were more willing to use unguided web-based self-help (β=.22, 95% CI 0.08-0.35). There were no consistent significant predictors of willingness to use support groups. More detailed information is presented in Table S1 in Multimedia Appendix 1.

Most (n=85, 70%) respondents indicated they would like to receive additional information about at least one of these modalities: web-based self-help (n=78, 63%), bibliotherapy (n=66, 53%), or support groups (n=61, 50%). We explored demographic, clinical, waiting list, and attitude variables predicting participants’ responses to receive additional information. Overall, there were few strong predictors of information seeking for LITs (Table 3 and Table 4). Willingness to use the specific LIT was predictive of information seeking for web-based self-help (odds ratio [OR]=2.20, 95% CI 1.20-4.28) and bibliotherapy (OR=1.66, 95% CI 1.02-2.80). Relative to individuals in other practice settings, individuals in private practice were more likely to seek information for web-based self-help (OR=5.33, 95% CI 1.48-23.42). There were no consistent predictors of seeking information about support groups.

The aims of this study were to describe current waiting list practices (aim 1), describe providers’ attitudes toward digital and nondigital LITs for patients on their waiting lists (aim 2), and explore predictors of providers’ willingness to use digital and nondigital LITs and their decisions to learn more about them (aim 3). Most providers (n=94, 69%) endorsed keeping a treatment waiting list. Among those who said they do not, nearly half (n=19, 45%) reported scheduling patients in the “distant future,” for example, 2-3 months away. Thus, most (n=113, 83%) providers in this sample had an opportunity to use LITs with people waiting for treatment. However, fewer than 20% (n=13-20) reported having recommended books, apps, or support groups for patients on a waiting list. The majority of those who endorsed maintaining a waiting list noted the estimated waiting time for their patients to access treatment was high and had increased since the COVID-19 pandemic. The difference in the average estimated waiting time currently and prior to the pandemic was on average about a month, with a significant increase, suggesting patients have experienced an increased delay in accessing mental health care since the onset of the pandemic.

This study has a number of limitations to consider. First, the generalizability of the study results may be limited due to the relatively small sample size and the representativeness of our study sample, which consisted primarily of CBT-oriented providers. However, our sample was relatively similar to the gender and racial or ethnic demographic characteristics of the psychology workforce in the United States [35], and CBT theoretical orientations have become the most popular among providers [36,37]. Additionally, it is possible that providers who are more interested in digital or nondigital LITs may have been more likely to participate in the study, resulting in a biased rating of attitudes relative to the population of mental health providers. Provider estimates of waiting times may likewise be biased or inaccurate in other ways. Previous research on waiting times for psychological services suggests that average waiting time ranges from 2-3 weeks to 2-3 months [38,39], but this varies across studies. Future studies could explore alternative methods, including the “secret shopper” methods, to get more accurate data on waiting lists and estimated waiting times. Lastly, this study assessed providers’ limited attitudes toward the listed interventions, specifically their perceptions of the intervention efficacy, accessibility, and their willingness to use it. Future research should investigate more detailed perceptions of these interventions through, for example, qualitative interviews to identify specific barriers and facilitators to digital and nondigital LIT implementation and use.

Despite its limitations, this study has notable strengths. To our knowledge, this is the first survey of providers to assess attitudes toward digital and nondigital LITs for patients on a waiting list. Additionally, beyond assessing attitudes, we also provided an opportunity to learn more about digital and nondigital LITs and assessed participants’ decisions to request additional information. Most providers were not currently recommending digital or nondigital LITs for patients on their waiting list, but attitudes toward the interventions were neutral to positive. Most were willing to use at least one intervention for patients on their waiting list. We found no evidence that providers had more positive attitudes regarding digital versus nondigital interventions. Generally, guided interventions were seen as more effective but less accessible than unguided interventions, which, in turn, were seen as less effective but more accessible. Together, these findings support our proposal that dissemination and implementation of digital and nondigital LITs (eg, bibliotherapy or DMHIs) while patients are on waiting lists could be a promising strategy to reduce the burden of untreated CMDs.

Although we found few significant predictors of attitudes and information seeking, we found that practice setting (ie, private practice vs other settings) was a predictor of attitudes and behaviors toward LITs. Individuals in private practice were more willing to use unguided bibliotherapy compared to individuals in other practice settings but were less willing to use guided bibliotherapy and unguided web-based self-help. Interestingly, at the end of our survey, individuals in private practice were more willing to learn about web-based self-help resources compared to individuals in other practice settings. These findings may reflect a knowledge gap wherein individuals in private practice currently do not perceive themselves to know which DMHIs to turn to, a barrier reported by other mental health providers [27,40,41], and hence they are less willing to use them. In addition to private practice setting, CBT theoretical orientation was predictive of willingness to use unguided web-based self-help. This finding may reflect the recent increase in iCBTs in research and practice settings, many of which are self-guided [42,43]. Increasing knowledge about digital and nondigital LITs and disseminating the interventions that individual providers are willing to use may be a useful way of increasing the reach of LITs [44]. One strategy to increase the use of LITs on treatment waiting lists would be to target the dissemination and implementation of digital and nondigital LITs to individuals who already have positive attitudes toward them—in our study, CBT practitioners and people with private practices. Alternatively, researchers could study interventions to increase the willingness to use digital and nondigital LITs by those not predisposed to using them, for example people who are not CBT practitioners.

We investigated treatment waiting lists and attitudes toward LITs for patients in waiting lists. Most providers appear to keep a waiting list, but most of them do not provide LITs to individuals on their waiting lists. In general, attitudes toward using LITs for patients in waiting lists were positive. Future research should investigate manipulating attitudes toward digital and nondigital LITs as well as structural barriers that may influence their use. For example, regarding individuals in private practice, who may be less likely to recommend guided LITs, qualitative data from our participants and from other studies [45] highlight questions about providers’ legal and ethical liability related to giving LIT guidance for participants on their waiting lists (eg, “I think one might assume a level of risk if participating in guided exercises but not therapy associated with their office”). Additionally, providers may be more, rather than less, likely to recommend LITs to patients with different features (eg, less severe symptoms) [46]. The perceived efficacy of an intervention also seems to be a major correlate of its use, so interventions that educate providers about LIT research are also worth exploring. Future work should clarify the nature of liability when recommending digital and nondigital LITs, as this may be an obstacle to uptake.