The pilot study was a prospective 1-year observational study, evaluated via a multimethod research design [15]. The participants answered questionnaires, height and weight were measured, and interviews with dropouts were conducted.

Patients referred for assessment and treatment at the obesity unit at the Örebro University Hospital, which is located in an urban area of Central Sweden, were invited to participate. Those who wanted to participate in the program were consecutively included. Referrals to the obesity unit of patients with obesity can be made by health care professionals, such as general practitioners or public health nurses. At the obesity unit, psychologists, dietitians, physiotherapists, and physicians are employed. In this study, the intervention was delivered by all staff, except the physician, and the providers are referred to as therapists throughout the paper. The study ran from November 2019 to January 2021.

The inclusion criteria were patients aged 18 years or older and either a BMI of ≥30 kg/m² or a BMI of 28‐29.9 kg/m² along with concomitant hypertension, prediabetes, type 2 diabetes, coronary artery disease, hyperlipidemia, hepatic steatosis, sleep apnea, or polycystic ovary syndrome. The exclusion criteria were severe mental illness, an ongoing eating disorder, pregnancy, breastfeeding, cancer in active treatment, weight loss treatment in the last 6 months, active abuse of alcohol or drugs, and being unable to express oneself in Swedish in speech and writing.

As part of the study, 3 individual visits to the obesity unit were included, before participation in the IDT-O, after completion of treatment at 6 months, and at the 12-month follow-up. The IDT-O started within 2 weeks of the first visit. Dropouts were interviewed after they had terminated their participation.

Participants were provided with access to the IDT-O program via a secure login to the Swedish national health e-services called 1177.se. This platform is used for digital treatment and for digitally based contact with the official health care system in Sweden. The program lasted for 6 months and comprised 12 treatment modules with participants working for 2 weeks per module. The main content of the modules is shown in Table 1. The modules primarily consisted of text, but videos and images were also included. Each module ended with 1 or more practice tasks to be completed before the next module was started. Each participant had regular written contact with 1 of 4 therapists at the obesity unit via the treatment platform; the same therapist provided written feedback on completed practice tasks and answered questions from the participant.

Participants who dropped out of the treatment were invited to participate in a telephone interview about their experiences with the treatment and their reason for dropping out. Prior to the interviews, the participants received oral information about the aim of the interviews. A semistructured interview guide was used, covering areas such as previous experience of digital working methods; experience of the modules of the IDT-O; amount of time needed to complete the modules, including the practice tasks; experiences of their digital contact with the therapist; and whether they had experienced any effects of their participation in the IDT-O. All interviews started with a question about their reason for dropping out. The questions were based on both clinical and scientific interest in the further development of the IDT-O.

The semistructured interviews were conducted by a researcher with nursing background experienced in qualitative methods (first author [AAC]), who had no previous relationship with the participants or the staff at the obesity unit. The duration of the interviews varied within 18‐69 minutes (mean 31, median 24.5 minutes). Interviews were digitally recorded, after consent from the informants, and were transcribed verbatim by a professional transcriptionist. There was a variation in time span from patients dropping out from the intervention and being interviewed. This was partly because it was not always explicit when a participant had decided to discontinue the intervention. Sometimes they just stopped fulfilling their tasks and did not reply to the therapists’ messages. The longest interval between dropping out and being interviewed was therefore just less than 4 months.

The mean and SD are presented for continuous variables, and relative frequencies are provided for categorical variables. The effect size (ES) of a change between baseline and 6- and 12-month follow-up was estimated by calculating the standardized response mean (SRM)—that is, the mean change divided by the SD of change. ES or SRM was evaluated according to standard criteria: <0.20=trivial, 0.20‐0.49=small, 0.50‐0.79=moderate, and ≥0.80=large [21]. The sample size was too small for significance testing of the treatment outcomes. Analysis was performed using SAS (version 9.4; SAS Institute Inc).

The analysis was inspired by the steps described by Graneheim and Lundman [22]. First, the first author (AAC), who both conducted the interviews and is experienced in qualitative methods, listened to all the interviews, corrected any errors from the transcriptions, and attempted to get a sense of the whole. Next, the interviews were imported into NVivo software (QSR International) for further analysis. Meaning units were identified and labeled with a code illustrating the content. Thereafter, the codes were scrutinized for similarities and differences and were clustered into categories describing the manifest content of the interview data. Finally, the categories were reflected on, and their latent content was identified in the form of subthemes and a main theme. The analysis was performed by the first author (AAC) and, in order to ensure rigor and trustworthiness, was later discussed with the other authors. The other authors all had experience of working with patients living with obesity, working as a licensed psychologist (JK), a general practitioner (SJ and MG), or a public health nurse (AI). During the process of analysis, the authors practiced reflexivity by discussions comparing their preunderstanding with what was revealed in the dataset [23].

This study was approved by the Swedish Ethical Review Authority (registration number 2019‐04695). At the first visit to the obesity unit, participants were informed orally and in writing about the study, and written consent was obtained. No compensation was offered to the participants, but the intervention and the health care visits included in the study were offered free of charge. The quantitative and qualitative data were deidentified before analysis. When presenting quotes in the “Results” section, this was done cautiously, ensuring that the information could not be traced back to a certain participant. All quotations have been marked with the participation number, instead of names.

A total of 20 participants started the IDT-O program and 35% (7/20) completed all 12 modules. Of the participants, 17 were females, the mean age was 44.2 (SD 16.4) years, and about half (11/20, 55%) had a university education (Table 2). Mean body weight was 117.1 (SD 19.8) kg, and mean BMI was 42.8 (SD 6.4) kg/m². Ten (8 females) out of 13 dropouts agreed to participate in an interview. Two participants could not be reached, and 1 declined participation. The majority of the participants had previous experience with weight-reducing interventions. There was variation in the number of modules participants completed before dropping out.

Of those who completed the 6-month treatment, the mean percent weight reduction was 2.1% (95% CI 0.1‐4.0) after 6 months and 2.0% (95% CI −1.1 to 5.1) after 12 months (Table 3). Two participants achieved a weight reduction of at least 5% after 12 months.

In the completers, dietary habits scores indicated a moderate improvement (SRM 0.50, 95% CI −0.31 to 1.28) after 6 months and a small improvement (SRM 0.25, 95% CI −0.51 to 1.00) after 12 months (Table 3). The change in self-reported physical activity showed a moderate improvement (SRM 0.56, 95% CI −0.26 to 1.35) after 6 months and a large improvement (SRM 0.93, 95% CI −0.08 to 1.87) after 12 months.

No change in OP distress scores was noted after 6 months (SRM 0.17, 95% CI −0.92 to 0.58), while a small improvement (SRM 0.43, 95% CI −0.37 to 1.19) was observed after 12 months. The change in OP avoidance scores indicated a large improvement (SRM 0.89, 95% CI −0.03 to 1.75) after 6 months and a moderate improvement (SRM 0.67, 95% CI −0.18 to 1.48) after 12 months.

The responses at the 12-month follow-up to the question of whether the treatment had helped in changing participants’ lifestyle habits were as follows: 42.9% (3/7) “a little,” 28.6% (2/7) “pretty much,” 14.3% (1/7) “a lot,” and 14.3% (1/7) “very much” (Table S1 in Multimedia Appendix 1). The answers to whether the treatment had positive effects on participants’ health were as follows: 28.6% (2/7) “small effects,” 42.9% (3/7) “fairly large effects,” and 28.6% (2/7) “great effects.” The same response pattern was obtained for the question about positive effects on well-being.

In completers, the mean treatment satisfaction rating for the 12 modules was 4.5 (range 3.7‐4.9, where 1 = Poor and 5 = Very good), while the mean rating for the relevance or meaningfulness of the practice tasks was 4.6 (range 4.0‐5.0, where 1 = Not relevant and 5 = Very relevant) (Table 4). The dropouts’ mean satisfaction ratings for the completed modules and the practice tasks were about the same as those of the completers. Dropouts completed on average 4.4 modules with a range of 1-9 modules.

The analysis resulted in 1 main theme—namely, The programme was OK, but it does not suit everyone—which was the latent content of all the data running through the categories and subthemes. Despite having dropped out from the IDT-O, the participants mainly believed that the content was relevant and of good quality; nevertheless, it did not suit them—at least not at this specific time in life. The 3 subthemes focus on 3 main messages. The first of these, It wasn’t for me, highlights issues explaining why the IDT-O did not suit some participants at this stage of life. The second subtheme, There were good things, points to aspects that were found to be valuable. The third subtheme, There are things to improve, comprises ideas that are worth considering in order to further improve the IDT-O. Table 5 provides an overview of the results.

The subtheme There were good things focuses on the participants’ positive comments about the IDT-O, comprising the categories The content is relevant for people like me, Internet treatment offers freedom, and The therapist was supportive.

Although the participants were positive to a certain degree, they also had insights about possible drawbacks. The categories in this subtheme are Parts of the programme and the digital platform were not user-friendly and The supportive aspect should be further developed.

This pilot study was a first step in evaluating the feasibility and effectiveness of an IDT program for individuals with obesity. Both qualitative and quantitative methods were used to assess the intervention. The findings indicate that the IDT-O holds potential as a treatment option for people with obesity, although only 35% (7/20) of the participants completed the program. The quantitative and qualitative findings show that the participants were generally satisfied with the content and design of the intervention. Those who completed the IDT-O lost some weight, reported improved dietary habits, increased physical activity, and improved psychosocial functioning, and maintained these behaviors for 6 months after completing the treatment. However, the quantitative findings were based on only 7 participants who completed the IDT-O and should be considered preliminary. The qualitative findings, defined in the 3 subthemes: “It wasn’t for me,” “There were good things,” and “There are things to improve,” also revealed that the intervention was not deemed useful by everyone, and various reasons for dropping out were provided. The interviewees also offered suggestions that may be useful for program improvements.

The efficacy of a digital intervention is linked to user motivation and engagement, with nonadherence being a common problem [24]. Studies show that almost half of those who start iCBT treatment do not complete the program [25,26]. In this study, 65% (13/20) of the participants dropped out of treatment, typically early in the program, indicating that the IDT-O program did not suit all participants. Some participants had unmet expectations for the treatment; for example, they wanted more detailed advice, more contact with the therapist, or had difficulty concentrating and reading. Some dropouts were due to factors external to the treatment, such as unforeseen circumstances in life that affected participation. This study was conducted during the Covid-19 pandemic, which may have contributed to the high dropout rate due to participants’ own or family members’ illness. Aside from this factor, the reasons for dropout identified in this study are broadly consistent with the findings of a qualitative study on nonadherence to iCBT for generalized anxiety disorder [24], as well as with patients’ and therapists’ experiences of challenges related to other internet-delivered interventions [9,27,28]. Given the issue of nonadherence, this finding underscores the importance of thoroughly assessing patients’ expectations and clarifying the goal of the program before offering such treatment [4], although doing so can be challenging [9]. Such assessments are crucial, both from a professional standpoint and for the patients, as both parties invest resources in the treatment. Additionally, patients who do not complete the treatment may experience feelings of guilt and shame, as the treatment is largely based on their own initiative [29,30].

Both the dropouts and the completers reported high satisfaction ratings regarding the completed modules and practice tasks, although this comparison was limited to the initial 5 out of 12 program modules. This positive view of the program content was supported by the findings based on interviews with the dropouts. The completers demonstrated improvements in self-reported dietary habits, physical activity, and psychosocial functioning at follow-up, in line with previous research [13]. The completers also stated that the IDT-O program had helped them change lifestyle habits and had positive effects on health and well-being. However, due to the limited number of participants who completed the program, the observed changes are uncertain and must be confirmed in a more comprehensive study. Interestingly, the dropouts also reported that they had benefitted from the content of the modules they had completed and had made some changes to their dietary habits.

The weight loss among completers 12 months after starting treatment was modest (–2.2 kg) and slightly lower than that observed in previous studies. In a comprehensive review including 44 studies on web-based obesity interventions, the average weight loss was –3.5 kg (range –0.1 to –7.8 kg) after treatments lasting between 2 and 24 months [12]. The modest weight loss might be explained by the fact that the content of the IDT-O program is not exclusively focusing on weight loss, but on change of behaviors, which is not a quick fix. From both a public health and individual perspective, even moderate weight loss is considered relevant [31,32], and improvements in dietary habits and increased physical activity after treatment can have further long-term positive effects on weight and health [33]. To increase the effectiveness of weight management, integrating the IDT-O program with pharmacological treatment for weight loss may facilitate the achievement of more substantial long-term weight reduction and weight maintenance thereafter. In a recent publication, the American Diabetes Association recommended that pharmacological obesity treatment should be used in combination with other therapies, focusing on nutrition, physical activity, and behavioral changes, to achieve health goals. They base their recommendations on research showing that addition of pharmacological treatment to lifestyle interventions improves weight reduction and weight loss maintenance [34].

Obesity is a stigmatized condition [33], and reducing weight stigma was one of the goals of the IDT-O program. In the study, the OP scale was used to measure stigma-related impacts of obesity on psychosocial functioning, a key domain in the assessment of QoL in people with obesity. Among the completers, improvements in psychosocial functioning (distress and avoidance) were noted at follow-up, likely due to a combined effect of weight loss and behavioral treatment. Reducing body weight and weight stigma can be complementary goals in obesity treatment. By focusing on reducing internalized stigma, beneficial effects on general health and QoL may be achieved, independent of weight reduction [33,35].

In this intervention, only written online communication between the therapist and the participant was used. The interviewed participants expressed a need for increased contact with the therapist and suggested that face-to-face or telephone contact should be included during the treatment—a finding consistent with other qualitative studies of iCBT [24,28,36]. Blended treatment, which combines online communication with face-to-face meetings, has been proposed as a means of enhancing iCBT interventions [7]. The use of video or phone consultations as supplements to written online communication is another option. However, the potential benefits of blended treatment are largely unknown, and further research is needed to determine the most effective mix of online and face-to-face consultations and whether or not the therapeutic alliance is increased by such modifications. Moreover, the potential benefits must be weighed against the increased costs of a blended intervention.

Some participants found it difficult to adhere to the standardized 2-week schedule for each module and suggested that the treatment should be more flexible and tailored to individual needs. This is a common reaction to internet-delivered therapy [24,28,37]. There were many reasons given for not managing the time limit of the scheduled modules. One that might need extra consideration was that the amount of text could be overwhelming for those with reading difficulties or having problems concentrating. The notion of extensive text to read [27] and the requirement for reading and writing skills [24] call for future innovative and assessable solutions, including text-to-speech features.

Due to the incomplete adaptation to mobile devices, participants encountered difficulties accessing the program via their phones. Given that mobile phones are the most common gateway to the internet today [38], it is critical for IDT programs to be fully optimized for mobile devices. As in the current findings, technical issues are a recurring topic when investigating patients’ experiences with IDTs [4,9,29]. These issues should be addressed when offering these treatments to patients or when developing IDTs, as technical issues have been given as a reason for dropping out of treatment [27].

The IDT-O program was evaluated in a real-life clinical setting without extra resources, which is a strength with regard to its further development and future implementation. However, the therapists were inexperienced with IDT, which may have affected the results of this study.

The use of qualitative and quantitative methods, along with the inclusion of data from both completers and dropouts, resulted in comprehensive data, which allowed us to evaluate different perspectives. The findings of this pilot study are valuable for designing a full-scale study and for further refining the IDT-O program. The trustworthiness of this study was enhanced by the fact that the interviewer was a neutral party, uninvolved in the design of the IDT-O program and with no prior or ongoing therapeutic relationship with the interviewees. In addition, the mix of the researcher’s professional backgrounds and preunderstanding contributed to the credibility.

Despite the strengths of the study, there are also some limitations to consider. First, the sample size was too small for significance testing; instead, we calculated effect sizes (ESs) to describe changes over time for those who completed the IDT-O. However, the ES estimates are uncertain because they are based on a small number of participants. Second, the dropout rate was higher than expected, but most of the dropouts participated in the qualitative interviews and shared their experiences, adding valuable insights to the evaluation of the IDT-O program. Third, the Covid-19 restrictions may have affected participants’ daily life habits regarding physical activity and eating habits [39], which in turn may have influenced the outcomes of this study.

In many respects, the IDT-O can be deemed feasible as a treatment option for people with obesity. Among those who completed the IDT-O, body weight decreased and dietary habits, self-rated physical activity, and psychosocial functioning improved. Despite these positive findings, the dropout rate was high, with participants indicating different reasons for discontinuing the treatment. Therefore, more research is needed before concluding the positive results. The results can, together with existing literature on the topic, form a base for future development and further evaluation of the IDT-O. Suggestions for further development of the IDT-O include adding a text-to-speech feature, improving the platform’s compatibility with mobile phones, and enhancing its overall accessibility and usability. In the future, we recommend a more tailored assessment of potential patients before offering the IDT-O, further development of the technical aspects of the platform, an evaluation of the benefits of adding individual visits with the therapist, and testing of the program when combined with other treatment regimens (eg, pharmacotherapy) for obesity. These modifications could potentially lead to greater effects on various aspects of patients’ health.