Since the introduction of positive psychology in 1998 [1], the proportion of published studies on this topic has grown rapidly each year compared to the entire field of psychological research [2]. Researchers have argued that the study of positive psychology is more capacious than the field of psychopathology because its primary focus is not on dysfunction but on well-being and optimal functioning [3]. Recent evidence indicates that positive mental health and mental illness are related but different continua [4-6]. This so-called two-continua model implies that individuals can experience positive outcomes such as life-satisfaction, meaning, and personal growth when mentally ill, and that the absence of mental illness does not automatically imply high levels of positive mental health [7,8]. As a consequence, researchers have underscored the importance of enhancing positive mental health in addition to preventing and treating mental illness, in clinical populations as well as in the general population [3,9].

A central aim in positive psychology is to increase the amount of flourishing worldwide [1,10]. Flourishing is defined as the presence of high levels of the affective or “feeling good” dimension of well-being (ie, hedonic well-being) in combination with high levels of the psychological functioning or “living well” dimension of well-being (ie, eudaimonic well-being) [4,9,11]. Compared to “languishers” (ie, people at the lowest levels of hedonic and eudaimonic well-being) and individuals with depression, research has shown that flourishers experience the least emotional distress and psychosocial impairment, have hardly any lost work days, and are most likely to survive [4,8]. Keyes and colleagues [12] have found initial support for the hypothesis that an increase in well-being protects against the development of mental illness. More specifically, they found that the risk for having a mental disorder was equally lowest for individuals who were flourishing in 1995 and in 2005, and for individuals who were flourishing in 2005 but not in 1995 [12]. The same study also indicates that it is possible to enhance the amount of flourishing [12]. However, there seems to be significant room for improvement because epidemiological studies using the same operationalization of flourishing have found that there were only 17% flourishers in the United States [4], 20% flourishers in South Africa [13], and 37% flourishers in The Netherlands [14]. Some researchers theorize that even small improvements in the level of well-being in the general population could generate large preventive effects on the amount of psychopathology [9,15]. Yet, little is known about how the amount of flourishing can be increased. The present study will evaluate a theory- and research- based positive psychology intervention (PPI) on efficacy, specific working mechanisms, and cost-effectiveness for enhancing well-being and flourishing in the general population.

A large number of theories have been used to develop interventions and exercises within the positive psychology movement, as can be found in the Key Competence Happiness Database [16]. Two meta-analyses have shown that PPIs significantly enhance well-being and alleviate depression in the short term, although effect sizes were relatively small [17,18]. These small effect sizes might be a result of the nature and intensity of the included interventions, which were mostly single-component interventions (ie, one or more individual exercises targeting one component of well-being). Examples are “the three good things exercise” or “savoring” targeting positive emotion, and “using one’s strengths in new ways” to increase flow. However, since well-being is a multifaceted construct [10,19] it might be more efficacious to promote well-being and flourishing with a multicomponent intervention that contains a variety of evidence-based individual exercises targeting 2 or more theoretically relevant hedonic and eudaimonic well-being components. For example, the well-being theory of Seligman [10] states that the components of positive emotion, engagement, relationships, meaning, and accomplishment (PERMA) are all necessary for flourishing. Also, multicomponent interventions are higher intensity, target more resources of well-being in an individual, and are often of longer duration than single-component interventions, which may increase their efficacy.

In the last decade, some multicomponent PPIs have been investigated, including different face-to-face interventions. Examples are (1) Positive Psychotherapy, a structured 14-session program developed for individual or group therapy in clinical settings based on Seligman’s well-being theory [10,20,21], (2) Well-being therapy, an individual psychotherapeutic strategy of 8-12 sessions that is based on Ryff’s model of psychological well-being [19] and therefore contains strategies and therapeutic techniques to influence the components self-acceptance, environmental mastery, positive relations with others, personal growth, autonomy, and purpose in life [22], and (3) the Working for Wellness Program, a structured 6-week group-based program for employees with a strong focus on strengths, flow, and social relationships [23]. The efficacy of these multicomponent PPIs on mental illness and positive mental health have been demonstrated in a wide variety of studies—including randomized controlled trials (RCTs)—although most of these studies used small samples [20,22,23]. However, the need for trained therapists in these intensive face-to-face PPIs is a disadvantage that may limit their feasibility, cost-effectiveness and dissemination.

Positive psychology exercises have the advantage of being relatively small and easy-to-implement in daily life, and are therefore well-suited for self-administered interventions. Recently, some Web-based multicomponent PPIs have been developed to increase the reach of PPIs. For example, Schueller and Parks [24] selected 6 exercises from the Positive Psychotherapy intervention (active-constructive responding, gratitude visit, life summary, three good things, savoring, and strengths) and randomly assigned participants to a 6-week program containing all 6 exercises, 4 of the exercises, or 2 of the exercises. The 2- and 4-exercise programs resulted in significantly larger effects on depressive symptomatology than the control group and the 6-exercise program, although participants in the latter program practiced the exercises on significantly more days than participants in the 2- or 4-exercise programs [24]. Another, more comprehensive intervention is Psyfit. This intervention contains 6 modules, each with 4 lessons, about goal setting, positive emotion, positive relations, mindfulness, optimism, and mastery [25]. An RCT showed significant but small effects on well-being and depressive symptomatology, although most participants completed less than 4 lessons of 1 or more modules [25]. While these and other Web-based multicomponent PPIs have shown that they can reach many individuals, their cost-efficacy seem to be limited by low adherence rates [24-27], which could be due to the lack of human contact and coaching [27,28].

Another, possibly cost-effective approach could be to intertwine the advantages of the aforementioned formats by using an accessible self-help book in combination with email support to reach and stimulate large groups of individuals at minimal costs. Meta-analyses have shown that the use of self-help books—also called bibliotherapy—with or without support are effective in the treatment of depression [29,30] and the treatment of alcohol problems [31]. Also, research on guided self-help in the form of email support in combination with face-to-face therapy or online therapy has shown promising results [32,33]. Besides, there are some indications that tailored email support is more effective than no support or automated email support [32,34]. More important, promising findings have been found for the use of positive psychology-based self-help books without any support [35-37] or with email support [38]. For example, Parks and Szanto [37] compared a PPI group of students who received the book “The How of Happiness” [39] to a group of students who received a cognitive behavioral self-help book to cope with depression and a control group. Results showed that both intervention groups were equally effective in reducing depressive symptoms at post-test compared to the control group, and that significant increases in life-satisfaction were found only for the PPI compared to the control condition at the 6-month follow-up. Another study used a multicomponent self-help book based on acceptance and commitment therapy in a sample with mild to moderate depressive symptomatology [38]. One intervention group received the book with extensive and tailored email support on progress and process, while another group received the book with minimal email support on progress. Results revealed significant increases in well-being and mindfulness and significant decreases in depressive, anxiety, and fatigue symptomatology for both intervention groups at 3-months follow-up compared to a waiting list control group [38]. In this study, we will use a version of the extensive email support similar to that used in the latter study [38]. That is, participants will be guided through the self-help book by tailored email feedback from trained counselors, but we will compare it to receiving an information email instead of minimal email support.

This study will examine a comprehensive multicomponent positive psychology-based self-help book with email support. The self-help book is entitled, “This is your life” (TL) [40] and targets 6 key components in positive psychology: positive emotion, use of strengths, optimism, self-compassion, resilience, and positive relations. We will compare a condition wherein participants receive the book with weekly asynchronous and tailored email support (TL-E) to a condition wherein participants receive the book with a weekly email with information (TL-I). Both groups will be compared to a waiting list control condition (WL). Our research aim is to evaluate the efficacy of the multicomponent self-help book with or without email support in terms of well-being and flourishing in an RCT (NTR4297) and to evaluate which of the included well-being components in the self-help book contribute to its efficacy. We expect that TL-E will be equivalent or superior to TL-I and superior to WL at post-test and 6 months follow-up, and that the effects will be sustained at 12 months follow-up.

This study will add to current knowledge in 3 ways. First, it will be conducted in a sample with low or moderate levels of well-being without moderate or severe clinical symptomatology, which is a potential “happiness-seekers” group [41] that has not previously been studied in PPIs [17,18,27]. Second, this study will gain insight into the working mechanisms underlying the intervention by evaluating which of the included well-being components in the self-help book (ie, positive emotion, use of strengths, optimism, self-compassion, resilience, and positive relations) contribute to its efficacy and by examining individual differences in demographic characteristics, number of life events, personality traits, and adherence rates. Third, this study will examine if the innovative format used in this study (ie, self-help book with relatively inexpensive email support) is a cost-effective approach for enhancing well-being and flourishing compared to a waiting list control group, since little is known about the cost-effectiveness of PPIs [26].

This study is a parallel RCT with 3 groups: a group receiving the self-help book with weekly asynchronous and tailored email support (TL-E); a group receiving the self-help book without email support but with a weekly information email (TL-I); and a waiting list control group (WL). Outcomes will be assessed at baseline (T0), post-test (around 3 months after baseline; T1), and at 6 (T2) and 12 months (T3) after baseline. All measurements will be self-reported and gathered via email with a link to a personal questionnaire built with Qualtrics software [42]. The design and procedures of this study were approved by the Ethics Committee of the University of Twente in The Netherlands (no. 13212).

Participants will be recruited from the general population in The Netherlands using newspaper advertisements. The recruitment messages will be positively framed: (1) “Get the best out of yourself and improve your resilience and well-being with a free self-help course”, and (2) “Become happy and stay happy? Improve your resilience and well-being with a free self-help course”.

Interested participants will be directed to a website with extensive information about the study and an application form. After applying, participants will be sent a personal link to a screening questionnaire on their specified email address. Participants who open the personal link will be asked to give online informed consent. If no informed consent is given, participants will be diverted to the information website and will not be able to participate in the study. If informed consent is given, participants will fill out the screening questionnaire before the start of the study. Figure 1 shows the intended flow-chart of participants.

Adults aged 18 or above with low or moderate levels of well-being without moderate or severe clinical symptomatology are eligible for this study because previous studies have shown that these individuals might search for happiness [41] and could boost their mental health by elevating the level of well-being [7,8]. Other inclusion criteria are a sufficient internet connection, a valid email address, and a willingness to invest an average of 4 hours per week for 8-12 weeks. In the screening questionnaire, the Mental Health Continuum Short Form (MHC-SF) will be used to assess the level of well-being and the Hospital Anxiety and Depression Scale (HADS) will be used to assess the level of clinical symptomatology. Participants will be excluded from the study if they score 4 or 5 (range 0-5) on at least 1 of the 3 emotional well-being items in combination with a score of 4 or 5 (range 0-5) on at least 6 of the 11 social and psychological well-being items of the MHC-SF, which is operationalized as flourishing (ie, high levels of well-being) [13,43]. Participants will also be excluded if they score 10 or higher (range 0-21) on the depression scale and/or on the anxiety scale of the HADS, which indicates individuals with moderate or severe clinical symptomatology [44]. Participants who will be excluded from the study will be notified by email because, (1) they are already flourishing, or (2) they probably have at least moderate depression or anxiety symptoms. The latter group will receive their specific scores and their meaning on the HADS and will be advised to talk to their general practitioner when symptoms persist or increase. After screening, the remaining participants receive a personal email link to the baseline questionnaire.

Randomization will be conducted after all eligible participants complete the baseline questionnaire, with 396 participants being randomly assigned to 1 of the 3 groups (allocation ratio, 1:1:1). Randomization will be stratified by gender and educational level (ie, low, medium, high) using a computerized random number generator created with Excel. This is an automated process with no interference from the investigators.

This study will use the self-help book, “This is your life” [40] in both experimental conditions. This book is mainly based on the overarching and comprehensive well-being theory of Seligman [10] and Ryff’s theory of psychological well-being [19]. The book consists of 8 modules on 6 key components in positive psychology: positive emotion; discovering and using strengths; optimism and hope; self-compassion; resilience and post-traumatic growth; and positive relations (Table 1). Each module contains psycho-education derived from specific theories and empirical evidence in positive psychology and related research areas. Each module also contains a number of positive psychology exercises, ranging from 3-10 per module. Most of these exercises are evidence-based, such as the "three good things exercise" [17,18] and "imagine your best possible self" [45-47]. All modules are, in theory, effective independent strategies for enhancing well-being. Participants in the experimental groups will receive the self-help book by regular mail, accompanied by a time schedule for reading the book and practicing the recommended exercises (see Table 1). Participants will be instructed to read one module per week—except for module 2, which can be spread out over two weeks—in sequential order, and to practice the recommended exercises of that module. If the participant has more time, he or she can re-read the module and practice other exercises of that module. Participants will be encouraged to invest the most time in the exercises that feel most beneficial to them. Participants will have 8-12 weeks to complete the 8 modules of the book to take holidays and other circumstances into account. When participants in the TL-E group need more than 9 weeks to complete the book, in consultation with the counselor, they will not receive email support in the weeks they are not in the position to work on a module. Participants in the TL-I group will receive weekly information emails during the first 9 weeks and will be asked to read the emails in chronological order in the weeks they actually work on a module. On average, participants in the experimental groups will be expected to invest 4 hours per week during the intervention period.

For a brief overview of all outcome measures and assessment times, see Table 2. Because one of our aims is to examine the working mechanisms underlying the intervention, which requires assessing all included positive psychology components, measures are selected on brevity and availability in the Dutch language.

Well-being is the primary outcome used for the power calculation. We expect to detect a significant effect on well-being between the experimental groups TL-E and TL-I compared to the WL group. Based on recent meta-analyses [17,18] and a previous comparable study [38], we expect to find a standardized effect size on well-being of 0.40 or larger. The targeted sample size is 99 participants per group (297 in total) to provide a statistical power of (1 - β) = 80% and a 5% significance level (two-tailed). To allow a study dropout rate of 25%, we need 132 participants in each condition and 396 participants in total for the trial.

The Consolidated Standards of Reporting Trials (CONSORT) statement [91] will be used to report the results. Independent t tests and Chi-square statistics (X²) will be used to examine baseline differences between the study groups. Non-significant socio-demographic differences at baseline will indicate successful randomization. Internal consistency of the constructs will be determined using Cronbach’s alpha. To examine the efficacy of the intervention, intention-to-treat analyses will be conducted by including all participants in the analyses who have been randomized. The Expectation Maximization (EM) method will be used to impute all missing data on the continuous measures of T1, T2 and T3. This method imputes the data by maximum likelihood estimation using the observed data in an iterative process [92]. The intention-to-treat analysis will be compared with a completers-only analysis.

To examine significant differences between the 3 groups, we will perform repeated measures ANOVA in a 3 (group) x 3 (time) design with well-being (MHC-SF and FS) as the primary dependent variable and positive/negative emotions (mDES), the use of strengths (SUS), optimism (LOT-R), self-compassion (SCS-SF), resilience (BRS), and positive relations with others (SPR) as the secondary dependent variables. Significant time x group interactions will be tested with Tukey’ post hoc tests. Within-group effects (Cohen’s d) will be calculated by subtracting the mean post-test or follow-up score from the mean baseline score and dividing the difference by the pooled standard deviation. Between-group effects (Cohen’s d) will be calculated by subtracting the mean post-test or follow-up score of the experimental group from the mean post-test or follow-up score of the control group, and dividing the difference by the pooled standard deviation [93]. These effect sizes are used to gain insight into the relevance of the significant effects. To examine changes in the amount of flourishers, languishers, and moderately mentally healthy (MHC-SF), descriptive statistics will be used. Changes in categories will be tested with X²-statistics.

Mediator and moderator analyses will be performed as described by Hayes [94,95]. Mediator analyses will be conducted to investigate which of the specific components of the intervention—positive emotion, use of strengths, optimism, self-compassion, resilience, and positive relations—contribute to its efficacy and mediate overall well-being. Moderator analyses will be conducted to examine which specific subgroups will improve most (or least) on well-being and flourishing from participation in one of the experimental groups. The independent variables of gender, educational level, life-events, personality traits, level of adherence to the intervention, and participation satisfaction will be entered as moderators in regression analyses.

A cost-effectiveness analysis will be conducted from a societal perspective. For each condition, the mean annual health care costs (MCQ), costs due to productivity losses (PCQ), and intervention costs will be calculated on an annual per capita basis for the reference year 2014. All costs will be expressed in Euro (€). The incremental cost-effectiveness ratio (ICER) will be calculated using a bootstrap approach. Then, a cost-effectiveness acceptability curve will demonstrate the probability of the cost-effectiveness of TL-E and TL-I compared to WL, for a range of willingness-to-pay (WTP) ceilings. All tests will be two-sided and will use alphas of 0.05. All data-analyses will be performed with SPSS version 21.0 or higher.