A standard ballroom and Latin dance program to improve fitness and adherence to physical activity in individuals with type 2 diabetes and in obesity

Lifestyle changes aiming at healthy diet and habitual physical activity are mandatory for the prevention and treatment of type 2 diabetes [1], as well as in other metabolic disorders [2]. Dietary restrictions and physical activity promote weight loss, reduce the progression from prediabetes to diabetes [3], improve metabolic control and reduce the long-term risk of complications, at any stage of disease severity. Most guidelines recommend 150 min/week of moderate intensity physical activity, corresponding to 10–15 MET-hour/week and no more than 2 consecutive days without exercise [1, 4]. Unfortunately, most adult individuals with metabolic disorders lead a sedentary life [5], and increasing physical activity is particularly challenging for overweight people due to aversion and poor tolerance of exercise [6].

Behavior changes are driven by personal motivation and stage of change to healthier lifestyles [7], and an increased intrinsic motivation for physical activity remains the strongest predictor of long-term results [8]. Specific programs to increase motivation and adherence to physical activity have been developed in subjects with type 2 diabetes [9], following the seminal experience of the Diabetes Prevention Program [10]. Both the metabolic control and the cardiovascular risk profile improve in parallel with the increased participation in structured physical activity programs [11, 12]. Nonetheless, procedures and strategies derived from motivational interviewing and cognitive-behavior therapy are rarely applied outside research settings, and most patients continue to receive suboptimal or ineffective treatments.

Exercise adherence tends to increase when patients are free to select their own program and less structure is imposed [13]. The adoption of pleasant programs of leisure-time physical activity may be a reasonable option to increase motivation and adherence [14], while maintaining a supervised control on the amount of physical activity. Among leisure-time activities, dancing has a remarkable place [15]. When arranged by patients’ associations, aerobic dance, Latin dance, folk dance and standard ballroom dance may all help aging people to enjoy their physical activity. Dance exercise also stimulates positive emotions, promotes social interaction and creates relationships with other people in the group, while exchanging experiences about their common medical problems. Finally, the acoustic stimulation and the music might strengthen the beneficial effects of aerobic exercise on cognitive functions [16].

We aimed to test the metabolic and clinical effects of a 6-month program of dancing in subjects with type 2 diabetes and/or obesity, chronically cared for in two diabetes/metabolic units. Self-selection of the leisure time activity program was allowed to increase adherence to the physical activity programs.

In general, the participation to DP was good (77%), with 34/41 attending >70% of planned sessions. One patient in DP and four in SSAP dropped out before the 3-month control; two patients more were lost in SSAP during the following three months (6-month attrition: DP 2%, SPP 10%, P = 0.233, Fisher’s exact test). Both physical activity programs produced a significant decrease of body weight (−2.6 kg for both; P < 0.001) and waist circumference (DP, −3.2 cm; SSAP, −2.2; P < 0.01) at 3 months, which were maintained at 6 months. Fasting glucose and liver enzymes decreased in both groups. Among biochemical variables, only glycosylated HbA1c showed a different time trend. At 3 months it decreased on average by approximately 0.2-0.3% in the two groups; in the following three months HbA1c decreased by an additional 0.1% in DP, whereas in SSAP it returned to the pre-study values.

No subject experienced hypoglycemic events.

Physical activity, energy expenditure, and 6-min walk test

The distance at 6MWT increased at 3 months in both groups; after 6 months, the distance further increased by 4% in DP and was unchanged in SSAP. The monthly time course of activity-related energy expenditure was significantly different between groups (ANOVA, P <0.001; Figure 1). In DP it averaged 13.5 ± 1.8 MET-hour/week in the first 3 months and 14.1 ± 3.0 in the second 3-month period. In SSAP it was characterized by a higher and highly variable expenditure in the first 3-month period (16.5 ± 13.9 MET-hour/week), which decreased thereafter to 14.2 ± 12.3 (P vs. first three-month period <0.001).

The study identifies the benefits of a supervised dance program as part of leisure-time physical activity in improving metabolic control and physical fitness in type 2 diabetes. DP was as effective as SSAP (home, gym, or open-air activity) in the short-term, but the benefits of SSAP were not maintained, with a higher attrition and a progressively reduced energy expenditure along the months, in spite of similar stage-of-change and motivation to physical activity at entry. This underlines the importance of social support and pleasant activities to increase adherence in individual patients, according to cultural and social heritage, translating into significantly better health and psychosocial targets.

The clinical benefits of both activity programs extended from body weight loss to improved glucose control and reduced liver enzymes, to improved physical fitness. The modest but significant reduction of liver enzymes is likely to reflect decreased hepatic triglyceride accumulation [30], as commonly observed in nonalcoholic fatty liver disease entering specific activity programs, also independent of weight loss [31, 32]. It should be viewed in the context of the chronic liver disease associated with obesity and diabetes, where intense behavioral treatment [33], also leading to improved glucose control, reduced insulin levels and improved insulin sensitivity [34], is expected to stop liver disease progression [33, 35].

The total amount of physical activity was probably underestimated in DP and overestimated in SPP. Most DP patients added physical activity bouts to the two weekly dance sessions, but these events were purposely not recorded to let patients completely free of implementing other activities according to their personal interests, while focusing on strict adherence to DP. On the contrary, all bouts of leisure-time activities were registered in SSAP as primary goal of the study, with a well-known risk of over-reporting compared with objective assessment [36]. Walking, biking and jogging were the most common activities selected in SSAP, with peaks of over 40 MET-hour/week in 7/54 SSAP cases.

During each DP session, the amount of activity was calculated as 4.3 MET-hour on the basis of a repetitive, predetermined Latin and ballroom dance program. According to the Compendium of Physical Activity [20], the energy dance expenditure may vary from 3.5 MET (Caribbean dance, beginners [37]) to >10 MET (Aerobic step and Swedish folk dancing) [38, 39]. The program was designed for mature/elderly people, and the theoretical maximum expenditure was limited between 4 and 7 MET [40], to avoid the risk of cardiovascular complications, without differences between genders [41]. This moderate amount was very similar to the amount of supervised aerobic and resistance training tested against physical activity counseling in a large randomized Italian study [12], where exercise significantly improved metabolic parameters. Similarly, 60-min brisk walking three times weekly (approximately 10 MET-hour/week) improved oxygen uptake, lipid and glucose homeostasis and reduced visceral fat [42]. All studies involving gyms, however, need an intense support to limit attrition, whereas the social support of dancing was probably pivotal to keep patients with diabetes on treatment, reducing HbA1c to an extent similar to that achieved in SSAP.

The estimated MET amount was on average >10 MET-hour/week. Di Loreto et al. have shown that physical activity above this target produces metabolic and clinical effects, which become large above the 20 MET-hour/week threshold [11]. This higher amount of physical activity, also associated with reduced costs of pharmacological treatment [11], was more frequently attained in SSAP, where individual cases attained weekly physical activity levels >40 MET-hour/week, but the minimum of 10 MET-hour/week was more common in DP, both during the first and, particularly, during the second 3-month period and the last month when 93% of cases (39/42) in DP met the target vs. 54% in SSAP (28/52; P <0.0001, Fisher’s exact test, in the per-treatment analysis, corresponding to 90% vs. 48% on intention-to-treat).

The study has both strengths and limitations. The main strength is its easy implementation in the community with limited resources and the robust results that might be exported to other typical clinical settings. The limits are the selection of motivated individuals – with minor differences between groups –, the non-randomized nature of the study, the so far time-limited experience, and the methods used to estimate energy expenditure. The selection of subjects after motivational interviewing may be justified by the very low motivation to physical activity commonly observed in individuals with type 2 diabetes [43], also reported in Italian subjects with diabetes [23] as well as other metabolic disorders [24], with a high risk of attrition in unselected patients. The lack of randomization was justified by inherent difficulties in directing individuals to dancing independently of their willingness. To adjust for non-randomization, a propensity score approach was selected. Although less solid than properly planned RCTs, adjusting by propensity produces valuable data reflecting the “real world” of disease treatment [44], and the procedure is largely accepted in chronic diseases requiring motivation as pre-requisite to accept treatment [44]. The study design was chosen to accommodate the patients’ desires to receive their favorite treatment and achieve physical activity benefits while maintaining scientific integrity. In this case, we tried to favor patients by providing a DP, which was also planned according their cultural heritage. In the real setting, every effort should be made to offer motivated patients their preferred activities, as long as they demonstrate long-term effectiveness, and also low-grade activity may be effective [42]. Forcing patients, although motivated, to perform activities largely considered unpleasant in the general population is likely to result in attrition and treatment failure [45]. As to the time-limited experience, the study will be continued and further work is needed to confirm longer effectiveness. Finally, the methods used to estimate energy expenditure has probably led to an underestimation of physical activity in DP and overestimation in SSAP. To overcome this limit, future studies should use more objective devices (i.e., accelerometers, not usually available in standard clinical settings) to assess participants’ expenditure.

In conclusion, dance may be an effective strategy to implement physical activity in motivated subjects with type 2 diabetes or obesity. Following this seminal positive experience, dancing might also be tested as a prevention strategy in cases at high risk of type 2 diabetes in the community, where physical activity may halt or delay disease progression [46]. As a social and pleasant form of exercising, it may be potentially useful, particularly to older and obese patients, also according to cultural and social heritages, to escape isolation and improve quality of life.