Erasmus MC heeft ervoor gezorgd dat je Mijn BSL eenvoudig en snel kunt raadplegen. Je kunt je links eenvoudig registreren. Met deze gegevens kun je thuis, of waar ook ter wereld toegang krijgen tot Mijn BSL.
Om ook buiten de locaties van Erasmus MC, thuis bijvoorbeeld, van Mijn BSL gebruik te kunnen maken, moet je jezelf eenmalig registreren. Dit kan alleen vanaf een computer op een van de locaties van Erasmus MC.
Eenmaal geregistreerd kun je thuis of waar ook ter wereld onbeperkt toegang krijgen tot Mijn BSL.
Login
Als u al geregistreerd bent, hoeft u alleen maar in te loggen om onbeperkt toegang te krijgen tot Mijn BSL.
The objective of this research was to map the Sussex Mindfulness Meditation (SuMMed) model to US adults, thereby demonstrating its utility for future researchers. Additionally, the research explored engagement with meditation beyond what the model explicitly prescribes, including with apps. This served as a means to offer potential refinements to the model and deepen our understanding of meditation habit formation.
Method
This research used a representative survey of 2000 adults in the USA. The sample was rigorously recruited using the Ipsos KnowledgePanel.
Results
In mapping the SuMMed model, 69.7%, 19.1%, 2.3%, and 9.0% of US adults were in the Pre-Intention, Preparation, Action, and Maintenance Stages of meditation habit formation, respectively. In the Preparation stage, 34.5% planned to meditate within the next 7 days, while only 21.9% intended to start a regular meditation practice within the next month. Additionally, people with less mature meditation practices were more likely to intend to use meditation apps, compared to established meditators.
Conclusions
The SuMMed model offers a useful way to understand meditation among US adults and future researchers should leverage it confidently. However, people often tried meditation out of curiosity and appeared to distinguish between trying and experimenting with meditation, versus having a practice. Given this, we recommend a modification to the Preparation stage of the SuMMed model to integrate meditation trial and experimentation. Additionally, meditation apps may be more useful to support early meditation practices, compared to established practices.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Mindfulness meditation, often simply referred to as “meditation,” has been practiced for centuries. According to Buddhist texts, one of the primary reasons to practice mindfulness meditation is to “reduce suffering” (Koenig, 2024). Today, people’s motivations for meditating are similar. The most often cited motivations for having tried meditation are to reduce stress and anxiety, alleviate depression, manage emotions, increase resilience, and enhance overall happiness and life satisfaction, among other health-related reasons (Anderson et al., 2019; Burke et al., 2017; Cramer et al., 2016; Lam et al., 2023a, 2023b; Macinko & Upchurch, 2019; Pepping et al., 2016). Additionally, numerous studies have directly demonstrated the health benefits of meditation in both clinical and non-clinical settings (Galante et al., 2021; Goldberg et al., 2018; Luberto et al., 2018). These studies, among many others, further reinforce the benefits and substantiate people’s motivations to meditate. For these reasons, one could reasonably assert that meditation is a form of health behavior.
Stage theories are common in health behavior research and are typically used to outline the process by which health behaviors might form into habits. According to stage theories, people move between discrete levels, or stages, of engagement with a given health behavior. Among the most popular stage models is the Transtheoretical Model (TTM) (DiClemente & Prochaska, 1998), in which people initially begin in the Precontemplation stage, during which they are not aware of or not intending to take action regarding a given health behavior. Next, people move to the Contemplation stage, in which people intend to make a change within the next 6 months. The third stage is Preparation, during which people intend to take action to engage in the behavior in the near term, typically within the next 30 days. From there, the fourth stage is Action, at which point people have made specific, overt modifications in their lifestyle in the past 6 months. Finally, people move into the Maintenance stage, in which the behavior has largely become a habit, people strive to prevent relapse to past behaviors, and they are more confident they can continue the specific lifestyle changes they have implemented (Prochaska & Velicer, 1997). Additionally, in the TTM, the stages themselves are not considered completely linear and unidirectional, but rather people can move back and forth between stages as their habit builds, is sustained, or regresses (Prochaska & Velicer, 1997).
Stage theories have been directly adapted to the meditation context by Miles et al. (2023) in the Sussex Mindfulness Meditation (SuMMed) model (Fig. 1). The authors argued that mindfulness meditation is distinct from many other health behaviors because it requires significant effort, it is often most impactful over longer time periods, and the results are not always apparent to others (Miles et al., 2023). Therefore, meditation requires a specific, unique theoretical model.
Fig. 1
The Sussex Mindfulness Meditation (SuMMed) model (Miles et al., 2023)
×
In the SuMMed model, which is similar in construction to the Transtheoretical Model, there are four discrete stages in the habit formation process with meditation. The first stage is Pre-Intention, in which people are not engaging with mindfulness meditation and have no intention of doing so in the foreseeable future. In the second stage, called Preparation, people are not engaging with meditation, but intend to do so in the near future. The third stage is called Action, which is demarcated by having recently begun a meditation practice and intending to continue the practice into the future. The fourth and final stage is Maintenance, at which point the person has been meditating for longer and intends to keep doing so in the future. The time period in transitioning between Action and Maintenance is not explicitly specified, but the authors suggested it can range from 8 weeks to 6 months, or more (Miles et al., 2023).
The SuMMed model also splits the four stages into two phases. These first two stages, Pre-Intention and Preparation, are designated as the Intentional Phase, during which no direct engagement with meditation has occurred. The third and fourth stages, Action and Maintenance, are designated as the Behavioral Phase and are distinguished by people having taken action by directly engaging with a meditation practice. This distinction between the Intentional and Behavioral Phases is informed by the Health Action Process Approach (HAPA) (Schwarzer, 2008), in which there is a clear and important distinction between not having and having taken direct action in engaging with the health behavior, in this case, meditation (Miles et al., 2023).
These two phases are an important difference between the Transtheoretical Model and the SuMMed model. In the TTM, Prochaska and Velicer (1997) did not outline any rules or guidelines that are required of the Preparation stage that explicitly state that individuals must not yet have engaged in the respective health behavior in any capacity. Rather, the authors provide examples of potential actions one might take, such as purchasing a book, joining a course, or talking to a professional. However, there is no mention of trying or not trying out the behavior, perhaps as a form of experimentation. According to the TTM, the key distinction between the Preparation and the Action stages is demarcated by specific and overt lifestyle modifications, and that not all modifications of behavior count as action. In the SuMMed model, Miles et al. (2023) drew a clear delineation in the Preparation stage, which is immediately before people begin to take any action to actively engage in meditation. This explicit rule is likely the result of integrating the TTM and the HAPA to create a more robust stage model for meditation habit formation. Whereas the TTM is less specific in defining the behaviors that are comprised of being in the Preparation stage (Prochaska & Velicer, 1997), the HAPA is more explicit about the importance of delineating between pre-intentional motivation processes indicative of intention, and post-intentional volitional processes indicative of behavior change (Schwarzer, 2008). Ultimately, the constraints applied to the Preparation stage in the SuMMed model through overlaying the intentional and behavioral phases in the HAPA may have created too rigid a framework for meditation habit formation in that people are either not meditating at all, or they have started a practice and intend to continue the practice. This may also pose a challenge because many people begin meditating out of curiosity (Sedlmeier & Theumer, 2020) and may not be committing to a practice right away.
With this context, the primary purpose of the present study was twofold. First, the study used the SuMMed model as a framework for which to map meditation prevalence estimates, offering an early application of this new model of meditation habit formation. In doing so, this research offers a more nuanced understanding of meditation engagement among US adults, including exploring the differences in the sociodemographic composition of different stages. Second, the present study further tested the SuMMed model by exploring other levels of engagement with meditation beyond what the model explicitly prescribes as stages. These include measures such as awareness of meditation, having tried meditation, having meditated recently (i.e., within the past 7 days), and intent to meditate in the near future (i.e., in the next 7 days). Additionally, because meditation app usage is fairly common among meditators (Lam et al., 2023a, 2023b), the research also explored meditation app and streaming content usage, as well as future intent to use this technology across the stages of the SuMMed model. By exploring the nuances of the stages with this level of detail, the present research can offer some potential refinements to the SuMMed model to consider.
The present study offers a unique contribution to our knowledge about meditation habit formation in several ways. First, we offer an early application of the SuMMed model that demonstrates its utility for future researchers. Second, we offer recommendations to consider for refining the model, as future researchers who study meditation continue to explore and better understand meditation habit formation. Finally, we explore the role of meditation apps and streaming content within the stages of the model to further understand the role of technology in meditation habit formation.
Method
Participants
This research is composed of a representative survey sample of 2000 total respondents (2044 total respondents, unweighted) ages 18 and older who live in the USA (see Table 1 for the weighted and unweighted demographic composition of the sample).
Table 1
Demographic composition of sample
N unweighted
Percentage unweighted
N weighted
Percentage weighted
Total sample
2044
100.0%
2000
100.0%
Age
18–34
591
28.9%
594
29.7%
35–49
488
23.9%
466
23.3%
50–64
517
25.3%
472
23.6%
65 +
448
21.9%
467
23.4%
Education
HS grad or less
672
32.9%
764
38.2%
Some college/Assoc
535
26.2%
528
26.4%
Bachelor’s or higher
837
40.9%
708
35.4%
Gender
Male
1024
50.1%
980
49.0%
Female
1020
49.9%
1020
51.0%
Income
Under 50 K US$
534
26.1%
528
26.4%
50 K US$–99.9 K US$
569
27.8%
568
28.4%
100 K + US$
941
46.0%
904
45.2%
Race and ethnicity
White, non-Hispanic
1376
67.3%
1226
61.3%
Black, non-Hispanic
216
10.6%
242
12.1%
Hispanic
289
14.1%
350
17.5%
Other
163
8.0%
182
9.1%
Region
Northeast
371
18.2%
346
17.3%
Midwest
438
21.4%
410
20.5%
South
757
37.0%
772
38.6%
West
478
23.4%
472
23.6%
Procedure
The survey fieldwork was conducted in March 2024. Respondents were recruited using the Ipsos KnowledgePanel. The Ipsos KnowledgePanel is a rigorously recruited online panel that is representative of the US population. Specifically, members of the KnowledgePanel are randomly recruited through probability-based sampling. Address-based sampling (ABS) methods are used to ensure full coverage of all households across the country. For non-internet households, Ipsos provides internet access and a laptop for the completion of online surveys. Once household members are recruited for the panel, they are notified by email to participate in surveys that they are assigned (Ipsos, 2024).
This study leveraged the KnowledgePanel Omnibus, which is a weekly recurring survey that fields Friday through Sunday. The Omnibus is structured such that Ipsos clients can purchase individual questions within the larger recurring survey. Potential respondents are recruited to participate through email and the survey is hosted online. The survey sample itself is recruited to be representative of the US adult population, and as such there are no additional screening criteria required to participate in the survey beyond socio-demographic quotas.
When fieldwork is completed, Ipsos builds design weights to account for any differential nonresponse that may have occurred. This is included in their standard respondent-level data deliverable. For this study, geodemographic distributions for the population were obtained from the March 2022 supplement of the US Census Bureau’s Current Population Survey (U.S. Census Bureau, n.d.). Weights are based on gender (male/female), age (18–29, 30–44, 45–59, and 60 +), race/Hispanic ethnicity (White/non-Hispanic, Black/non-Hispanic, other or 2 + races/Non-Hispanic, Hispanic), education (high school or less, some college, bachelor, and beyond), census region (Northeast, Midwest, South, West), metropolitan status (metro, non-metro), and household income (less than US$24,999, US$25,000 to US$49,999, US$50,000 to US$74,999, US$75,000 to US$99,999, US$100,000 to US$149,999, US$150,000, or higher). An iterative proportional fitting (raking) procedure was used to produce the weights. The resulting weights were scaled to aggregate the total sample size of all eligible respondents. The rigor of this approach allows us the opportunity to project the data to the US adult population.
Measures
This questionnaire consists of 11 survey questions, developed for this study. In order to reduce cognitive load and limit self-report biases, the survey questions were designed to be concisely worded and inquire about recent time periods. The questions include awareness of meditation, whether respondents have ever practiced meditation, whether respondents have practiced meditation in the past 7 days, the number of days in the past 7 days they have practiced meditation, and their intention to practice meditation in the next 7 days. The survey also includes awareness of meditation apps and meditation content on streaming services, the number of days in the past 7 days that the person practiced meditation using an app or streaming service (which was grouped to create a variable regarding whether the person did or did not use a meditation app in the past 7 days), and intention to practice meditation in the next 7 days using a meditation app or content from a streaming service. The questionnaire also included a question to assess the specific stage in which the respondent self-identified as being within the current stages of the SuMMed model. Note that due to a survey programming change approximately halfway through fieldwork, the number of days in the past 7 days that the person practiced meditation using an app or streaming service is only available for about half of respondents.
For the purpose of this study, meditation was defined for participants as “In meditation, a person focuses, stills, or quiets the mind.” This specific framing was chosen for several reasons. First, a version of this question framing has been used in the National Health Interview Survey (NHIS), a highly rigorous study conducted by the US Center for Disease Control and Prevention (Centers for Disease Control & Prevention, 2022). Second, the wording is simple and concise, which is important to reduce non-response bias (Bauer et al., 2025). This framing has also been tested through cognitive evaluation (Smith et al., 2022). In this evaluation, while Smith et al. offered that some potential interpretations of meditation could appear out of scope (e.g., meditation for me is exercising, reading), this may not necessarily be the case because meditation and mindfulness can be integrated into everyday activities, which may in fact be an important goal of the practice. Additionally, the data from the present study on the prevalence of lifetime meditation exposure are in line with other research, at about half of respondents (Lam et al., 2023a, 2023b), indicating the present definition is viable (see Supplementary Table S1 for the questionnaire used for this research).
SuMMed Stages Model
The SuMMed model survey question includes each of the high-level categories within the model. For the analysis, those response option categories include the following: Pre-Intention (“I do not intend to start a regular meditation practice within the next 6 months”), Preparation (combining “I am thinking about starting a regular meditation practice within the next 6 months” with “I intend to start a regular meditation practice within the next month”), Action (“I have been practicing meditation regularly for less than 6 months”), and Maintenance (“I have been practicing meditation regularly for more than 6 months”).
Further Measures of Meditation Engagement
In exploring the additional measures of engagement with meditation of the SuMMed model, several measures are overlaid onto the model. Those measures include awareness of meditation, whether respondents have ever tried meditation, short-term meditation recency (whether respondents have meditated within the past 7 days), and short-term meditation intent (whether respondents intend to meditate in the next 7 days). The 7-day period was chosen because self-report accuracy is higher for shorter versus longer time periods (Roberts et al., 1996; Short et al., 2009) and because 7-day retention for meditation apps specifically is low (Baumel et al., 2019). In doing so, whether someone has meditated in the past 7 days could serve as an indicator of whether someone is successfully beginning to establish a meditation practice or whether their practice is potentially faltering.
Demographics
As part of the standard data deliverable for omnibus studies, Ipsos provides demographic data for each respondent, including age, highest level of educational attainment, ethnicity and race, gender, geographic region, and household income, among others.
Data Analyses
The data analysis for this study was conducted using Python. The weighted data was used when estimating population proportions. To create population estimates, the weighted proportions were multiplied by the July 2023 US Census Bureau population estimates of the number of adults ages 18 and older in the USA. This figure is approximately 262 million people (US Census Bureau, n.d.).
For any given question included in this analysis, 1.0% or fewer respondents skipped the question and subsequently provided missing data. The only exception is the question previously mentioned (the number of days in the past 7 days that the person practiced meditation using an app or streaming service), which was subjected to a programming change midway through fieldwork. Any missing response data was omitted on a question-by-question basis throughout the analysis, rather than removing the respondent’s answers entirely from the analysis.
To explore socio-demographic skews across the SuMMed model, multinomial logistic regression was used, employing the statsmodels Python package (see https://www.statsmodels.org/stable/). The dependent variable was the specific stage of the SuMMed model. The Pre-Intention stage was held out as the dependent variable reference category. Age, education, gender, geographic region, income, and race and ethnicity were used as independent variables. Age was coded as an integer by the age groups described in the “Measures” section, beginning at one for ages 18–24 and increasing for each subsequent age grouping. Similarly, education was coded as an integer, with “Less than high school” coded as one and increasing by one for each subsequent increase in educational attainment. Income was also coded as such, with “Less than 10,000 US$” in annual income coded as one and increasing for each subsequent increase in income bracket. Gender, geographic region, and race and ethnicity were one-hot encoded as dummy variables, with “Male,” “South” and “White, Non-Hispanic” held out as reference categories for each attribute, respectively.
Results
Meditation Engagement
Among adults aged 18 and older in the USA, approximately 92.0% are aware of meditation, 49.8% have ever tried meditating, and 15.1% reported having meditated in the past 7 days. The proportion of people who have meditated in the past 7 days among people who have tried meditating indicates that approximately 70.7% of people who have ever meditated have either failed to establish a practice, have a lapsed practice, or may be at risk of lapsing. Among people who meditated in the past 7 days, the average number of days meditated was 3.9 (SD = 2.16). Approximately 18.1% intend to meditate in the next 7 days.
For meditation apps and meditation content on streaming services, 65.2% are aware this content exists. Additionally, 4.1% have used meditation apps or consumed meditation content on streaming services in the past 7 days, and 6.6% intend to use a meditation app in the next 7 days. This suggests that about 27.2% of meditators use meditation apps and/or meditation content from streaming services to support their practice. Figure 2 provides the population projection estimates based on the various forms of meditation engagement.
Fig. 2
Population estimates for varying forms of meditation engagement. Population estimates are calculated by multiplying the weighted proportion from the survey by 262 million, the estimated number of US adults from the US Census Bureau. Error bars are based on a 95% confidence interval. Meditation apps and streaming content consolidated to “app” in x-axis labels. WAU is an acronym for “weekly active user” (single asterisk). This is commonly used in the technology sector and indicates the person has engaged in the activity in the past 7 days. Intention defined as planning to meditate in the next 7 days (double asterisk)
×
Mapping the SuMMed Model
When mapping the SuMMed model directly to the population, we estimate that among US adults, 69.7% are in the Pre-Intention stage of the model, 19.1% are in the Preparation stage, 2.3% are in the Action stage, and 9.0% are in the Maintenance stage (see Fig. 3 for the population projection estimates based on the stages of the SuMMed model).
Fig. 3
Population estimates based on SuMMed model stage. Note: Population estimates are calculated by multiplying the weighted proportion from the survey by 262 million, the estimated number of US adults from the US Census Bureau. Error bars are based on a 95% confidence interval
×
Socio-Demographic Skews Across the SuMMed Model Stages
In the multinomial logistic regression model, the Pre-Intention stage is the reference category for the dependent variable, so the socioeconomic skews are based on comparisons to that category.
In the Preparation stage, race and ethnicity have the strongest skews in relation to the Pre-Intention stage, with people who identify as Black the highest (OR 2.33, p < 0.001), followed by people who identify as two or more races (OR 1.87, p < 0.05) and Hispanic (OR 1.83, p < 0.001). Females are more likely to be in the Preparation stage than males (OR 1.65, p < 0.001). Additionally, people with higher levels of education are more likely to be in the Preparation stage compared to the Pre-Intention stage (OR 1.22, p < 0.001). Finally, age has an inverse relationship with being in the Preparation stage, where younger segments are more likely to be in the Preparation stage than the Pre-Intention stage compared to older segments (OR 0.88, p < 0.001). The odds ratio for the constant for this stage in the model is 0.13 (p < 0.001).
Similar to the Preparation stage, in the Action stage, people who are Black have the highest odds ratio (OR 3.02, p < 0.01). Education level also shows a significant positive relationship with the Action stage (OR 1.42, p < 0.05). Additionally, income has an inverse relationship (OR 0.81, p < 0.05). No other variables are statistically significant in this stage. The odds ratio for the constant for this stage in the model is 0.34 (p < 0.001).
In the Maintenance stage, race and ethnicity similarly have the highest odds ratio, starting with people who identify as 2 + races (OR 2.48, p < 0.05), Black (OR 2.04, p < 0.01), or Hispanic (OR 2.00, p < 0.01). Additionally, education is correlated with being in the Maintenance stage, compared to the Pre-Intention stage (OR 1.18, p < 0.05). The odds ratio for the constant for this stage in the model was 0.05 (p < 0.001) (see Table 2 for additional details from the multinomial logistic regression modeling).
Table 2
Odds ratio (OR), coefficient, p-value, standard error, and from multinomial logistic regression of socio-demographic skews across the SuMMed model
Odds ratio
Coefficient
Std err
Preparation stage
Constant
0.131***
− 2.032
0.27
Age group
0.88***
− 0.127
0.035
Education level
1.223***
0.201
0.056
Gender identification
Female
1.651***
0.502
0.117
Male
-
-
-
Income group
1.015
0.015
0.038
Region
Northeast
0.82
− 0.199
0.174
Midwest
1.08
0.077
0.159
South
-
-
-
West
1.111
0.106
0.154
Race and ethnicity
2 + races, non-Hispanic
1.867*
0.624
0.296
Black or African American, non-Hispanic
2.33***
0.846
0.183
Hispanic
1.827***
0.603
0.171
Other, non-Hispanic
0.939
− 0.063
0.296
White
-
-
-
Action stage
Constant
0.034***
− 3.374
0.671
Age group
0.879
− 0.129
0.093
Education level
1.42*
0.351
0.15
Gender identification
Female
1.367
0.313
0.308
Male
-
-
-
Income group
0.81*
− 0.211
0.096
Race and ethnicity
Black or African American, non-Hispanic
3.017**
1.104
0.411
Hispanic
2.061
0.723
0.426
Other, non-Hispanic
1.184
0.169
0.76
2 + races, non-Hispanic
0.95
− 0.052
1.039
White
-
-
-
Region
Midwest
0.549
− 0.599
0.477
Northeast
0.773
− 0.258
0.426
South
-
-
-
West
0.739
− 0.303
0.415
Maintenance stage
Constant
0.05***
− 2.996
0.383
Age group
1.067
0.065
0.049
Education level
1.181*
0.167
0.078
Gender identification
Female
1.262
0.233
0.162
Male
-
-
-
Income group
0.955
− 0.046
0.051
Race and ethnicity
2 + races, non-Hispanic
2.483*
0.909
0.371
Black or African American, non-Hispanic
2.035**
0.711
0.26
Hispanic
2.003**
0.695
0.231
Other, non-Hispanic
1.423
0.353
0.363
White
-
-
-
Region
Midwest
0.789
− 0.237
0.241
Northeast
0.895
− 0.111
0.235
South
-
-
-
West
1.2
0.182
0.205
*p < 0.05; **p < 0.01; ***p < 0.001
Mapping Other Measures of Engagement to the SuMMed Model
Within the Pre-Intention stage of the SuMMed model, awareness is an important measure, given that people need to have heard the term and at least know what meditation is in order to progress to the Preparation stage. Awareness of meditation among people in the Pre-Intention stage is 88.5%. Roughly one-third (35.4%) of people in this stage have ever tried meditating. Awareness of meditation apps and content on streaming services is 59.5%.
Within the Preparation stage, 72.6% have ever tried meditating. When we explore the timeline during which people in the Preparation stage intend to meditate, 17.0% have meditated in the past 7 days, and 34.5% plan to meditate within the next 7 days. In this stage, 21.9% intend to start a regular meditation practice within the next month, and 78.1% are considering starting a regular meditation practice in the next 1–6 months. Note the discrepancy between people who plan to meditate in the next 7 days (34.5%) and people who plan to start a meditation practice in the next month (21.9%). For meditation apps and streaming content specifically, 76.7% are aware of these services, 8.2% have used them in the past 7 days, and 16.8% intend to use them in the next 7 days.
In the Action stage, we can assess meditation recency and short-term meditation intention to understand whether their practice could be waning, given that having a “regular” meditation habit is subjectively defined by respondents. Within this stage, 77.7% of people have meditated in the past 7 days. Among people who have meditated in the past 7 days, the average number of days they have meditated is 3.1 (SD = 2.14). Moreover, 90.2% intend to meditate in the next 7 days. This suggests that for some people, they still believe that they have an established practice, even if they may miss a week. For meditation apps and streaming content specifically, 81.7% are aware of these services, 38.0% have used them in the past 7 days, and 45.0% intend to use them in the next 7 days.
Finally, like the Action stage, we can explore meditation recency and short-term meditation intention in the Maintenance stage. Within the Maintenance stage, 86.6% of people have meditated in the past 7 days. Among people who have meditated in the past 7 days, the average number of days in the past 7 days they have meditated is 4.1 (SD = 2.55). Similarly, 91.2% intend to meditate in the next 7 days. For meditation apps and streaming content specifically, 81.0% are aware of these services, 20.1% have used them in the past 7 days, and 26.0% intend to use them in the next 7 days.
Post Hoc Exploration
When we discovered that meditators in the Maintenance stage appear to be less likely to intend to use a meditation app or streaming service than people in the Action stage (45.0% in the Action stage compared to 26.0% in the Maintenance stage), we wanted to explore further. In doing so, we conducted a logistic regression, using the intention to use a meditation app in the next 7 days as the dependent variable, then including the same independent variables listed in the previous model. Additionally, being in the Action or Maintenance stage of the SuMMed model (Action was coded as one, Maintenance coded as zero) and the number of days meditated in the past 7 days, coded as an integer, was also included. From this model, identifying as female had the strongest relationship with intending to use a meditation app or streaming content (OR 2.77, p < 0.01), followed by being in the Action stage (OR 2.72, p < 0.05). Age was inversely related to using meditation apps or streaming content (OR = 0.70, p < 0.001), indicating that younger people are more likely to use apps and streaming content to support their practice (see Table 3 for additional details on this model). The pseudo r-squared from this model is 0.13, indicating a significant amount of potential variance that is unexplained.
Table 3
Odds ratio (OR), coefficient, p-value, standard error, and from multinomial logistic regression of meditation intention
Odds ratio
Coefficient
Std err
Constant
0.13*
− 2.044
0.798
SuMMed model level
Action
2.719*
1.00
0.391
Maintenance
-
-
-
Number of days meditated, past 7 days
1.115
0.109
0.068
Age group
0.699***
− 0.358
0.097
Education level
1.311
0.271
0.158
Gender identification
Female
2.772**
1.02
0.334
Male
-
-
-
Income group
1.068
0.066
0.106
Race and ethnicity
2 + races, non-Hispanic
1.104
0.099
0.78
Black or African American, non-Hispanic
1.228
0.206
0.497
Hispanic
1.473
0.388
0.424
Other, non-Hispanic
0.584
− 0.538
0.843
White
-
-
-
Region
Northeast
0.901
− 0.104
0.469
Midwest
0.853
− 0.16
0.502
South
-
-
-
West
1.234
0.21
0.419
*p < 0.05; **p < 0.01; ***p < 0.001
Discussion
This research makes several contributions to our understanding of meditation. First, by rigorously mapping the SuMMed model to the US adult population using a probability sampling technique, we are able to better understand both the model itself and the dynamics of the US population. Using the additional measures of meditation engagement included in this analysis that are not a part of the SuMMed model, the analysis revealed that people within various stages of the model have substantive differences in their engagement with meditation. This validates the model as having internal consistency. Future researchers should feel comfortable leveraging the SuMMed model.
The research also uncovered that a substantial number of people who are in the Preparation stage (and do not yet have an established practice) report having meditated in the past 7 days (17.0%). Additionally, in the Preparation stage, 34.5% plan to meditate within the next 7 days, and 21.9% intend to start a regular meditation practice within the next month. Taken together, the results suggest that this is not simply people regressing from the Action stage to the Preparation stage in the SuMMed model, but rather that people see a difference between meditating and having a meditation practice. Further, the results indicate that people may try out or experiment with meditation before considering whether they want to attempt to establish a meditation practice. This is supported by Sedlmeier and Theumer (2020), who found that many people begin meditation, including both new and experienced meditators, simply out of curiosity. This suggests that experimenting with meditation is not an uncommon entry point into establishing a meditation practice. In our view, this necessitates a revision to the definition of the Preparation stage of the SuMMed model, which is defined by people not yet having engaged with meditation. We recommend removing the constraint in the SuMMed model that people in the Preparation stage have not yet engaged with meditation, given experimenting with meditation appears to be a common and potentially important aspect in developing a meditation habit.
Conversely, among people who reported having an established meditation practice (i.e., being in the Maintenance stage), a substantive number did not meditate in the past 7 days (13.4%) or may not intend to meditate in the next 7 days (8.8%), suggesting that people who believe they have an established meditation practice may not feel rigidly defined by the frequency with which they practice.
The research also uncovered that about twice as many people are in the Preparation stage compared to the Action and Maintenance stages combined. While it is unlikely that all of the people in the Preparation stage will actually establish a meditation practice, this is a strong indicator that there is a high level of interest in and potential for growth in meditation. Additionally, nearly one-third of people in the Pre-Intention stage have tried meditation, suggesting that meditation had captured their interest at one point in time, so the door may not be closed in progressing to the Preparation stage. However, more research is needed to understand how to break down the barriers to meditation for these individuals (e.g., low perceived benefits, perceived inadequate knowledge, perceived pragmatic barriers, and perceived sociocultural conflict, see Hunt et al., 2020; Williams et al., 2011) if they are to establish or reestablish a practice.
When we explored potential socio-demographic differences across the stages of the SuMMed model, race and ethnicity were particularly prevalent (e.g., identifying as Black, Hispanic, or 2 + races), compared to gender and education. This is in contrast to some past research, which found that meditation tends to skew non-Hispanic White and female (Burke et al., 2017; Cramer et al., 2016; Lam et al., 2023a, 2023b; Upchurch & Johnson, 2019). This may suggest that as meditation has grown recently in the USA (Davies et al., 2024), the practice may be attracting a more diverse population of people. Additionally, this research found that only the Preparation stage has a significant female skew and that the Action and Maintenance stages appear to be more balanced across people who identify as male versus female. However, like other research has found, these results further support that meditation app usage, specifically measured in this research by intention, has a female skew (Bhuiyan et al., 2021; Jiwani et al., 2023).
This research also found that about one-in-four meditators use meditation apps and/or meditation content on streaming services. Interestingly, people in the Action Stage are more likely to intend to use meditation apps in the future, compared to people in the Maintenance stage. This suggests that meditation apps and streaming content are more commonly used to support an early meditation practice, but once meditators have established their practice and understand meditation more deeply, this type of meditation support may not be as important or useful. Lam et al., (2023a, 2023b) found that doubts regarding app effectiveness were a barrier to usage, which established meditators may have experienced. Additionally, the cost of a subscription for meditation apps was the largest barrier to usage (Lam et al., 2023a, 2023b), and if experienced meditators found they were not all that effective, they may have subsequently decided the apps are not worth the cost. People in the Maintenance stage may also be less likely to intend to use apps and streaming content because people who have a more established meditation practice have integrated meditation more deeply into their everyday life and subsequently may not need apps, in which the meditation context is often seated practice. It could also be that content on meditation apps is more oriented toward beginning practitioners, but for people with an established practice the content is no longer new, insightful, or novel. This is a clear opportunity for future researchers to replicate and explore more deeply.
Limitations and Future Directions
There are several important limitations to this research. First, data are self-reported and therefore prone to inaccuracy, including when reporting health behaviors (Prince et al., 2008). Considering that people are also more accurate in their self-reporting of more recent behaviors (Roberts et al., 1996; Short et al., 2009), the questionnaire was structured in a way that erred toward more recent behaviors, but this approach leaves blindspots in exploring long-term behaviors. Additionally, while we briefly described what meditation is to respondents via the survey instrument, there is still a risk that conceptually it is ambiguous and there could have been potential confusion about what meditation is exactly.
Future research might more deeply explore the extent to which people try out and experiment with meditation before deciding whether to attempt to establish a meditation practice. Exploring behavioral log data with meditation apps may be useful in better understanding this phenomenon because it offers a nuanced perspective on actual behaviors. Additionally, it may be useful to better understand the dynamics of experimenting with meditation. Specifically, what motivates meditation experimentation, including the role of technology, and subsequently what further compels people to move from meditation experimentation to deciding to try to establish a regular practice.
Finally, future researchers may consider further substantiating that meditation apps skew toward newer meditators and more deeply understanding why this may be the case. While Lam et al., (2023a, 2023b) uncovered some important barriers to using meditation apps overall, such as costs and perceived effectiveness, more research may be useful to more deeply understand newer versus established practitioners’ experience with meditation apps and streaming content.
Acknowledgements
The study would not have been possible without Spotify. The data for this research was repurposed from a survey initially commissioned through Spotify.
Declarations
Ethics Approval
The study was approved by the Kent State University Institutional Review Board.
Informed Consent
Adults who join the Ipsos KnowledgePanel consent to be invited to surveys. Per the agreement Ipsos KnowledgePanel has with panelists, they are aware that every survey and every survey item is voluntary and the data collected and provided to Ipsos KnowledgePanel clients is anonymized.
Conflict of Interest
The primary author is employed by Spotify, which is a streaming service that offers music, podcasts, and audiobook content. Some content on Spotify could be considered meditation related.
Use of Artificial Intelligence
Artificial intelligence tools were not used for this research.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
share
DELEN
Deel dit onderdeel of sectie (kopieer de link)
Optie A:
Klik op de rechtermuisknop op de link en selecteer de optie “linkadres kopiëren”
Met BSL Psychologie Totaal blijf je als professional steeds op de hoogte van de nieuwste ontwikkelingen binnen jouw vak. Met het online abonnement heb je toegang tot een groot aantal boeken, protocollen, vaktijdschriften en e-learnings op het gebied van psychologie en psychiatrie. Zo kun je op je gemak en wanneer het jou het beste uitkomt verdiepen in jouw vakgebied.