Inadequate physical activity (PA) costs more than 100 billion dollars annually and leads to 10% of early deaths (U.S. Department of Health and Human Services [HHS], 2018). Physical activity decreases risks for noncommunicable diseases, cancers, dementia, and early death when practiced at a sufficient level over many years (HHS, 2018). The problem is that most individuals do not routinely engage in enough PA to meet national guidelines over enough time to reap these benefits and achieve healthy longevity (HHS, 2018).
The purpose of this project was to determine if a nurse practitioner (NP)–led program for PA habit formation would allow interested ambulatory clinic patients to form a habit of risk-reducing PA. Habits are less difficult to continue than to lapse from (Bargh, 1994; Hagger, 2019). Thus, development of a PA habit can lead to prolonged PA and better health (HHS, 2018). The aim of this report is to disseminate actionable information on an innovative interventional program for the formation of PA habits, which promote health and reduce risk factors. This theory- and evidence-based program was implemented as a Doctor of Nursing Practice project with a convenience sample of 100 nurse-patients from an NP-run clinic.
Available knowledge (background)
Benefits of physical activity
It is important to support patient development and continuation of PA routines meeting federal guidelines, as they enhance health, decrease disease risks, and lower health care expenditures. Despite the known benefits of decreased risk of morbidity and mortality from sufficient PA, most patients in the United States do not regularly get enough PA to obtain these benefits (HHS, 2018).
For self-insured employers, physically active employees incur lower health care costs as does the employer (Black, 2017). Monetary costs for health care decrease, but absenteeism also decreases and productivity increases (Black, 2017).
Contribution of habit to persistence of physical activity
Although no studies were found linking PA habit to long-term PA, theory indicates that habits are more difficult to break than to continue (Aarts, Paulussen & Schaalma, 1997; Bargh, 1994; Gardner, 2015; James, 1890). Thus, forming a PA habit should support the continuation of PA.
Habit theory is evolving but consistently proposes that habits predict continuation of routine PA (Aarts, Paulussen, & Schaalma, 1997; Gardner, Rebar, & Lally, 2019; Hagger, 2019; James, 1890). Repeated behaviors become habits when practiced routinely in a consistent context, which serves as a cue or impulse to act (Gardner et al., 2019). Habits take less thought to continue than to stop (Gardner et al., 2019). Thus, theoretically, a PA routine practiced in the same context, which serves as the cue to act, becomes a habit and persists over time.
Ryan's (2009) integrated theory of health behavior change (ITHBC) and habit theory of Aarts et al. (1997) framed the interventions. Each of these complex theories is briefly summarized below.
Ryan's (2009) ITHBC is a comprehensive theory integrating salient points from multiple behavior change theories. The major concepts in this theory begin with knowledge, beliefs, and social facilitation. These concepts have many components and are postulated to contribute to a person's ability to regulate their behavior. Thus, self-regulation is another major, multifaceted concept in the theory. It is postulated to lead to the ability to manage one's behavior long term, identified in the ITHBC as engagement in self-management behavior. Self-management of behavior is theorized to lead, over time, to the final major concept in the theory, that is, health.
Habit theory (Aarts et al., 1997), as it relates to PA, builds on earlier self-efficacy work and social learning theory of Bandura (1977). Aarts et al. emphasized that habits stem from personal choices based on motivation and social norms, which self-efficacy theory addresses. They also added that routine repetition is an important factor in habit formation. They postulate that, over time, satisfying repeated behaviors require less conscious decision making as memory of them persists. They also hypothesized that physical exercise behaviors, once habitual, are triggered in response to a situation (cue) and goal (e.g., jogging after breakfast).
Available evidence is sparse for PA habit development. Habit development was studied in a small group (n = 13) of participants, followed daily to determine the time it takes to develop a PA habit (Lally, Van Jaarsveld, Potts, & Wardle, 2010). Participants selected a PA they wanted to develop into a habit. They also selected a context to serve as a reminder or cue to begin the activity. Achievement of 95% of the stable habit level was reported to take 91 days in this study, which used the Self-Report Habit Index (SRHI) (Verplanken & Orbell, 2003).
One other study examined PA habit formation; this time in new gym members who were predominantly female college graduates who were overweight (N = 111) (Kaushal & Rhodes, 2015). Consistently engaging in PA and rewards for doing so were found to predict PA habit development in this small, short-term study. The researchers concluded that habit formation peaked at 6 weeks into the intervention. Those who engaged in exercise more than four times per week for 6 weeks and developed a PA habit, as measured by the Self-Report Behavioural Automaticity Index, persisted in their new habit over the remaining 6 weeks of the study (Gardner, Abraham, Lally, & de Bruijn, 2012; Gardner et al., 2019; Kaushal & Rhodes, 2015).
Short-term PA promotion projects generally do not lead to long-term persistence of enough PA to enhance health and result in risk reduction (Gardner et al., 2019; Gardner et al., 2012). However, PA persistence is the key to gain long-term health benefits, such as reduced risk of chronic diseases. Habit formation is an important process that leads to persistence of behavior change, as it is easier to continue a habit than to break it (Gardner et al., 2019; Hagger, 2019). The aim of this 3-month quality-improvement project was to achieve PA behavior change and PA habit formation among patients who completed the program.
A quantitative interventional design was used to determine whether an innovative, technology-based, intervention program promoted PA habit development and increased PA behavior in a convenience sample of employee health clinic patients. The principal investigator (PI) developed and conducted formative and summative evaluations to assess the feasibility for recruitment and acceptability of program procedures and retention of participants throughout the 3-month program. These were important considerations for expanding the program beyond the initial group of 100 nurses, to potentially offer the program to all 11,000 employees served by the system's employee health clinics.
The program was offered to all eligible nurse patients in two hospitals of a large health care system through an e-mail recruitment flyer. The first 100 respondents who came to the NP-managed employee health clinic for an intake visit to receive an explanation of the program, sign consent, complete intake measures, and commit to routinely engage in a cued PA of their choice for 3 months were enrolled in the program.
The program was developed based on behavior change and habit theories and available evidence (Aarts et al., 1997; Kaushal & Rhodes, 2015; Lally et al., 2010; Ryan, 2009) The feasibility and acceptability for employee-patients and health care system leadership were considered. Patient input supported the technological basis of the program, as it needed to be flexible to accommodate their work, family, and other commitments. Additionally, the program was structured to promote PA on break and other nonwork time based on the health care system's administrative input.
This interventional program was deemed a quality-improvement project by the college of nursing's Doctor of Nursing Practice Program Committee. The health care system's institutional review board agreed and approved before implementation.
A grant received from an international nursing society's chapter at the college of nursing defrayed some of the pedometer costs. In kind contributions from the health care system included staff time and technology assets.
The program staff consisted of registered nurses and NPs. All staff were educated on the program, intake, and follow-up processes by the lead NP running the program. Staff members completed at least one intake visit supervised by the lead NP before conducting intake visits independently. Ongoing monitoring assured fidelity to the planned program.
Following ethics and administration approval, the 3-month intervention began. A cap of 100 patients was placed on the initial program offering to allow for the evaluation and refinement of the program before increasing its scale. To assure fair selection of participants, all eligible nurses in the participating hospitals were e-mailed a flyer explaining the program, and the first 100 visiting the clinic and completing the intake visit were enrolled in the program. Participants' demographic characteristics are provided in Table 1.
Intake visits included an explanation of the program; informed consent; measurement of height, weight, and blood pressure (B/P); completion of the SRHI (Verplanken & Orbell, 2003); commitment to routinely engage in the patient's choice of PA in their chosen context to serve as a trigger or cue (e.g., walk in the park on the way home from work), log daily step counts, and text step counts to the NP daily. To improve knowledge and boost motivation, participants also agreed to read weekly educational and confidence-building e-newsletters throughout the 3-month intervention program and accepted occasional coaching texts. Additionally, participants agreed to return for follow-up biometric and habit measures at the end of the program.
The only restriction on the chosen PA was that it had to be measurable by step count for self-monitoring purposes, in line with the ITHBC (Ryan, 2009). Participants were each issued a research-grade pedometer (Yamax Digimax SW-200) to measure their steps, which they monitored and recorded daily for 3 months. Step-count goals were to increase daily step counts by an average of 200 steps/day or more on a weekly basis until a weekly average of 10,000 steps/day was reached and maintained (e.g., an average of 5,000 steps/day in week 1 led to a goal of at least 5,200 steps/day on average in week 2).
Reporting of step counts and scripted response texts
Step counts were texted daily to the NP program leader who responded when the counts were out of the individual's expected range for two or more consecutive days. Responses consisted of motivational scripted texts expressing concern, encouragement, or praise as warranted. If step count texts were not received for 2 days in a row, a scripted text was sent to prompt a report. Reporting was prompted no more than three times during the 3-month program. As multiple missed activity days could hinder formation of a habit (Lally et al., 2010), those who failed to report steps more than three times or did not report their steps for 7 consecutive days were considered to have dropped out of the program for evaluation purposes. However, they were allowed to continue to report step counts and receive text responses and e-newsletters.
Weekly newsletters were sent to each participant's chosen e-mail address. These e-newsletters were developed to both educate and inspire motivation to achieve risk-reducing levels of PA, in accordance with the ITHBC premise that knowledge and beliefs influence behavior change (Ryan, 2009). Weekly educational topics ranged from benefits of engaging in regular PA to advice on exercise equipment and safety. Inspirational e-newsletter messages were crafted to build confidence and promote can-do attitudes. Tips from other participants and prompts for making plans to overcome barriers to routine PA were also included (e.g., walk in a mall or exercise at the gym during inclement weather).
The intervention components are summarized in Table 2.
Self-report habit index
Habit was determined using the SRHI (Verplanken & Orbell, 2003), the most commonly reported habit measure. The SRHI has acceptable reliability and validity across multiple studies and behaviors with Cronbach alphas ranging from 0.85 to 0.95, depending on behavior and context; acceptable pretest–posttest correlation (r = 0.91; p < .001); and convergent validity (r = 0.58; p < .001) (Verplanken & Orbell, 2003). Habit formation for the participant's chosen PA was determined by pre- and postintervention scores on the SRHI.
Blood pressure measurements were taken pre- and postintervention. An aneroid sphygmomanometer was used for all B/P readings, which were taken manually by a registered nurse or NP.
Height, weight, and body mass index
Height and weight were measured using a manual Health-O-Meter scale. Height was measured in feet and inches at the intake visit and assumed not to change. Weight was recorded in pounds pre- and postintervention. Body mass index (BMI) was calculated pre- and postintervention by entering the height and weight of each participant in the online National Institutes of Health BMI calculator (HHS, 2018).
Steps were measured using a research-grade Yamax Digimax SW-200 pedometer issued to each participant at intake. Step counts were recorded on a log provided to the participant and self-reported by text daily to the NP program leader.
Descriptive statistics were used to characterize the sample. To evaluate pre- versus postintervention measurements, t-tests were run using SPSS version 24.
The sample included 100 nurses from two hospitals in a large health care system who volunteered to participate in response to an e-mail flyer sent to all eligible employees' work e-mail addresses (Table 1). The majority were female (96%), mirroring the gender composition of all eligible employees. Each participant desired to form a habit of routinely engaging in a PA of their choosing and identified a cue to prompt that behavior at intake.
More than half of participants completing the program developed a strong habit of PA (60.6%; 20 of 33). A decrease in habit scores by 25% (pre-SRHI = 62.94; post-SRHI = 46.90) and significant t-test score (t = 2.35; p = .025) indicated formation of PA habits on average by the completion of the program. The SRHI had a possible range of 12–132, with the cutoff for strong habit set at a score of 44 or below. All completing participants met PA guidelines for sufficient PA.
For those who completed the program, first week daily step counts averaged 7,867, although average step counts throughout the program for these participants were significantly higher at 9,313 (t = 14.421; p < .001). This is consistent with the theory-based framework inclusion of cued PA habit goal setting, routine performance of PA, self-monitoring entries in daily log, reporting step counts daily to the provider, receiving feedback to support self-management, and weekly e-newsletters for knowledge and confidence building (Aarts et al., 1997; Ryan, 2009).
Weight, body mass index, and blood pressure
Nearly half of all participants were overweight by BMI (42%). Nine participants had systolic B/P above 140 mm Hg at intake, none above 148. All nine were counseled on appropriate PA, hydration, and Dietary Approaches to Stop Hypertension (DASH) Eating Plan (HHS National Heart, Lung, and Blood Institute, n.d.) and scheduled follow-up at the clinic or with their private provider.
At the conclusion of the 3-month program, average weight decreased by 1 pound and BMI by 0.2 kg/m2. These values were neither clinically nor statistically significant (tweight = 1.47; p = .152; tbmi = 1.72; p = .096). This was as expected in this short program targeting increased PA, as routine PA is helpful for weight maintenance but has little to no effect on weight loss over short periods (Hall et al., 2011).
A summary of average results for those completing the program is presented in Table 3.
Feasibility and acceptability
Participant retention in the program was consistent with other PA programs, although not ideal. Feasibility for recruitment was acceptable. The maximum number of nurse participants (N = 100) completed intake measures within 1 week following an e-mail recruitment flyer sent to all nurses at the participating hospitals. Acceptability for completion of the program requirements of intake visit measures, daily texting of step counts, completion of the 3-month program, and return visit for follow-up measures was higher than expected at 33% (n = 33). Other PA programs have reported attrition rates as high as 90% or as low as 10%, the latter with monetary incentives for participation (Kramer, Tinschert, Scholz, Fleisch, & Kowatsh, 2018; Santa Mina et al., 2019; Vandelanotte et al., 2018). Please see Table 3 for a summary of results.
Physical activity habit formation was supported through this interventional program. Most participants who completed the 3-month program developed a habit of their chosen PA linked to a contextual cue. This behavior change to routine engagement in a self-chosen PA through the program supported the ITHBC propositions that knowledge and beliefs lead to self-regulation (motivation) ability, which leads to self-management behavior for health improvement (Ryan, 2009). In line with habit theory, PA habits formed through three months of repetition of cued PA in this program and are likely to continue, as habit processes make behaviors easier to continue than to lapse from (Aarts et al., 1997; Gardner, 2015; Gardner et al., 2019).
Retention in the 3-month program was 33%. This exceeded the benchmark set at 25% for the program. It is not uncommon for retention in PA programs of three months or longer to be 35% or less (e.g., Santa Mina et al., 2019; Vandelanotte et al., 2018). One Swiss study had much higher retention rates than all other studies reported (90.5% in one study arm), perhaps owing to the incentives offered or cultural differences. Anecdotally, it was not a part of the project measures but was noticed that participants who paired up or completed the program with a group (e.g., other nurses on their unit) finished the program. An implication for future research based on this observation is to modify the program to encourage dyads or teams to form at the start of the program and determine whether this is associated with higher retention. This is consistent with the ITHBC component of social facilitation supporting behavior change (Ryan, 2009). One implication for future research is to determine predictors of retention through PA program completion to determine what factors could support higher retention rates.
Weight and BMI did not change significantly during the 3-month intervention, as expected. Although most participants were overweight, weight loss was not an aim of this PA habit formation program. The program had a limited duration, and it occurred over the holiday season of November through January. Significant weight loss was not anticipated. However, a habit of sufficient PA supports weight maintenance, which is consistent with this program's results over the holiday season. Should participants choose to lose weight, an implication is to add a healthy eating component to the program, such as MyPlate or the DASH Eating Plan, which could result in weight loss (HHS, 2018; HHS National Heart, Lung, and Blood Institute, n.d.; U.S. Department of Agriculture, n.d.).
Blood pressure did not decrease significantly among participants. Because systolic B/P was 140 or above in only 9% of participants at baseline, and decreasing B/P was not an objective of the program, no significant decrease in B/P was expected. A longer period of sufficient PA could lead to normalization of B/P (HHS, 2018). Follow-up with participants at an additional times following the program could provide information on whether PA habit continued and B/P decreased for participants with hypertension.
Anecdotally, it was noted that participants were more likely to complete the program and follow-up visit if they joined with a group of work colleagues or had a friend participating in the program with them. It may be helpful to add support for forming exercise partner dyads or teams to future programs and determine whether friend participation predicts PA habit formation and persistence. This form of social facilitation is consistent with the ITHBC (Ryan, 2009).
Patients who volunteered to participate in this study all indicated a desire to develop a new habit of sufficient PA to enhance health, which may have influenced their results. Readiness for change may increase the likelihood of PA behavior change (Lundqvist, Börjesson, Larsson, Cider, & Hagberg, 2019).
Employee-patients who volunteered to participate in this program were a convenience sample with an affinity for the topic, both potential sources of bias (selection bias). Potential self-report bias is also a limitation, as participants texted their step counts and completed the SRHI (Verplanken & Orbell, 2003) as the habit development measure for this interventional program.
Although increased PA has led to increased productivity, decreased use of sick time, and increased morale in some studies (Centers for Disease Control and Prevention, 2017); these factors were not measured in this study due to its relatively short length.
The successful implementation and results of this technology-based innovative PA habit-development program indicate its feasibility and acceptability for patients in the employee clinic setting. The successful program supports the propositions of the theoretical frameworks—Ryan's (2009) ITHBC and Aarts et al.’s (1997) habit theory. Technology made the program convenient and cost-effective.
Research is needed on factors such as peer or friend support, which may improve retention in the program, as was anecdotally found in this project. Retention in program is important, as a retention period of 6 months to 1 year may increase the likelihood of formation of a strong PA habit for lifelong health (Lally et al., 2010).
Replication of the program at other types of practices is warranted, particularly for other primary care sites and oncology, cardiology, and endocrinology or diabetes education sites. This could leverage the health enhancing benefits of PA habits to improve or prevent hypertension, type 2 diabetes mellitus, and hyperlipidemia and reduce risks for diseases such as cardiovascular disease and cancer (HHS, 2018).
Acknowledgements:The author thanks Susan Bulfin, D.N.P., who served as faculty advisor and Lorrie Jones, Ph.D., who served as the worksite sponsor for this project. The author acknowledges Iota Xi Chapter of Sigma Theta Tau International for providing competitive grant funding and Memorial Healthcare System for in-kind contributions supporting this program.
Aarts H., Paulussen T., Schaalma H. (1997). Physical exercise habit
: On the conceptualization and formation of habitual health behaviours. Health Education Research, 12 363–374.
Bandura A. (1997). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall.
Bargh J. S., (1994). The four horsemen of automaticity: Awareness, efficiency, intentions, and control. In Wyer R., Srull T. (Eds.), Handbook of social cognition (pp. 1–40). Hillsdale, NJ: Lawrence Erlbaum.
Black A. (2017). Five reasons employee wellness is worth the investment. Office of Disease Prevention and Health Promotion. Retrieved from https://health.gov/news/blog/2017/05/five-reasons-employee-wellness-is-worth-the-investment
Centers for Disease Control and Prevention. (2017). Physical activity
: active people, healthy nation. Retrieved from https://www.cdc.gov/physicalactivity/index.html
Gardner B. (2015). A review and analysis of the use of “habit
” in understanding, predicting and influencing health related behavior. Health Psychology Review, 9, 277–295.
Gardner B., Abraham C., Lally P., de Bruijn G.-J. (2012). Towards parsimony in habit
measurement: Testing the convergent and predictive validity of an automaticity subscale of the Self-Report Habit
Index. Journal of Behavioral Nutrition and Physical Activity
, 9, 102–113.
Gardner B., Rebar A. L., Lally P. (2019). “Habitually deciding” or “habitually doing” A response to Hagger (2019). Psychology of Sport & Exercise
. doi:10.1016/j.psychsport.2019.05.008. Retrieved from https://www.sciencedirect.com/journal/psychology-of-sport-and-exercise
Hagger M. S. (2019). Habit
and physical activity
: Theoretical advances, practical implications, and agenda for future research. Psychology of Sport and Exercise
, 42, 118–129.
Hall K. D., Sacks G., Chandramohan D., Chow C. C., Wang Y. C., Gortmaker S. L., Swinburn B. A. (2011). Quantification of the effect of energy imbalance on bodyweight. Lancet, 378, 826–837.
HHS National Heart, Lung, and Blood Institute. (n.d.). DASH eating plan. Retrieved from https://www.nhlbi.nih.gov/health-topics/dash-eating-plan
James W. (1890). The principles of psychology (Vol. 1). New York, NY: Henry Holt and Company.
Kaushal N., Rhodes R. E. (2015). Exercise habit
formation in new gym members: A longitudinal study. Journal of Behavioral Medicine, 38, 652–663.
Kramer J.-N., Tinschert P., Scholz U., Fleisch E., Kowatsh T. (2018). A cluster- randomized trial on small incentives to promote physical activity
. American Journal of Preventive Medicine, 56, e45–e54. Retrieved from https://www.sciencedirect.com/science/article/pii/S0749379718322864
Lally P., Van Jaarsveld C., Potts H., Wardle J. (2010). How are habits formed: Modeling habit
formation in the real world. European Journal of Social Psychology, 40, 998–1009.
Lundqvist S., Börjesson M., Larsson M. E., Cider A., Hagberg L. (2019). Which patients benefit from physical activity
on prescription (PAP)? A prospective observational analysis of factors that predict increased physical activity
. BMC Public Health, 19, 482–494.
Ryan P. (2009). Integrated theory of health behavior change
: Background and intervention development. Clinical Nurse Specialist, 23, 161–170.
Santa Mina D. S., Au D., Auger L. E., Alibhai S. M. H., Matthew A. G., Sabiston C. M., Jones J. M. (2019). Development, implementation, and effects of a cancer center's exercise
-oncology program. Cancer, 125, 3437–3447.
U.S. Department of Agriculture. (n.d.). ChooseMyPlate. Retrieved from https://www.choosemyplate.gov/
U.S. Department of Health and Human Services [HHS]. (2018). Physical activity
guidelines for Americans (2nd ed.). Washington, DC: Author.
Vandelanotte C., Duncan M. J., Kolt G. S., Caperchione C. M., Savage T. N., Van Itallie A., Mummery W. K. (2019). More real-world trials are needed to establish if web-based physical activity
interventions are effective. BMJ British Journal of Sports Medicine, 53, 1553–1554. doi:10.1136/bjsports-2018-099437.
Verplanken B., Orbell S. (2003). Reflections of past behavior: A self-report index of habit
strength. Journal of Applied Social Psychology, 33, 1313–1330.