A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study.

BACKGROUND: Lack of physical activity (PA) and a high level of physical inactivity (PI) are associated with a higher risk for mortality and responsible for several non-communicable diseases including cardiovascular disease. Higher age is associated with a decrease of PA and an increasing level of PI. Studies have shown that interventions in the elderly have the potential to increase the amount of PA and to decrease the level of PI. However, most interventions are complex, elaborated, time- and resource-consuming. Here, we examined the effect of individual feedback-letters reporting the measured PA and PI in a sample of elderly people in Germany. Primary outcomes of the study were overall PA and PI after 6 months in the intervention group compared to a control group.
METHODS: We examined data from the MOVING intervention study (RCT) for people aged ≥ 65 years living in the northeast of Germany. At baseline, 3 and 6-months follow-up, all study participants wore a 3-axis accelerometer over a period of seven consecutive days. After the baseline measurement, the participants were randomized into intervention and control group. Participants in the intervention group received automatically generated, individualized feedback letters reporting their PA and PI by mail after the baseline measurement and after the 3-months follow-up. A Two-Way Mixed ANOVA with repeated measures was calculated with light, moderate and overall PA as well as PI as dependent variables, and group (between subject) and time (inner subject) as factors. The analysis based on retrospective data from the MOVING study (2016-2018).
RESULTS: N = 258 patients were recruited. N = 166 participants could be included in the analysis, thereof N = 97 women (58.4%). The mean age was 70.8 years (SD 4.8). At baseline, the participants had a mean wearing time of 5,934.5 minutes (SD = 789.5) per week, which corresponds to about 14 hours daily on average. The overall PA in the intervention group at the 6-months follow up was 2488.8 (95% CI 2358.9-2618.2) minutes and 2408.2 (95% CI 2263.0-2553.4) minutes in the control group. There was no statistically significant interaction effect (time*group) between the intervention and control group for the depending variables. Sensitivity analyses showed significant small positive effects of the interaction time*partnership, F(2, 300) = 3.020, p = 0.05, partial η2 = 0.020. DISCUSSION: On average, study participants had high levels of PA at baseline and showed a good adherence in wearing the accelerometer. Both is likely due to selection in the convenience study sample. Thus, some ceiling effect reduced the overall intervention effect somewhat. At baseline, the weekly average of PI was 3436.7 minutes, which correspondents to about 8.2 hours per day and about 57% of participants' daily waking time. The average level of PI could be slightly decreased in both study groups. TRIAL REGISTRATION NUMBER: DRKS00010410, 17 May 2017.

Introduction

There is strong evidence, that physical activity (PA) promotes healthy aging and that both lack of PA and a high level of physical inactivity (PI) are crucial risk factors for global mortality and several non-communicable diseases (NCD) [1-7].

PI is defined as “any waking behavior characterized by an energy expenditure ≤1.5 METs (metabolic equivalent of task) while in a sitting or reclining posture” [8, 9]. 31.1% of the worldwide adult population has a sedentary lifestyle, is insufficiently physically active and does not meet the recommendations for PA from the World Health Organization (WHO) [10]. Physical inactivity causes 3.2 million deaths per year worldwide. In 2010 PI was responsible for 69.3 million DALYs (disability-adjusted life years) [11]. PI causes a significant economic burden, in 2013, PI caused 53.8 billion US-Dollar healthcare costs worldwide [12]. As a consequence, the reduction of insufficient PA is a global target of the WHO [11, 13].

Depending on the definition, the proportion of PI across European countries ranges from 43.3% in Sweden up to 87.8% in Portugal. In Germany, 70.2% of men and 71.8% of women show a sedentary lifestyle [14]. High levels of moderate to vigorous physical activity (MVPA) are associated with a reduced risk for all-cause mortality [6, 15, 16]. In general, epidemiological evidence about positive effects of regular PA is strong and has accumulated over several decades [4, 17–19].

Several studies have shown that a physically active lifestyle (including promoting PA and minimizing PI) is associated with improved overall health status [4, 20] and a reduced risk for overall mortality by 22–34%, and CVD mortality by 27–35% [1]. In addition, regular PA is a well-established method to prevent diabetes, hypertension and several types of cancer (i.e. breast and colon cancer) also for older people. Furthermore, regular PA reduces the risk of developing stroke about 25–30% [21]. Even people who are only 1–2 times per week physically active, have a reduced risk for mortality [22] compared to those with less PA.

There is strong evidence, that moderate-to-vigorous PA (MVPA) has positive effects on important health outcomes and is associated with a reduced risk of all-cause mortality [16]. Several international physical activity guidelines from the United States [23], United Kingdom [24], Australia [25], and from the WHO [26] promote the benefits of both moderate and vigorous PA. Recommendations for light PA are lacking in the guidelines [24-27]. However, increasing light PA also seems to be positively associated with relevant health outcomes, i.e. obesity and glucose metabolism, and is associated with a reduced risk for all-cause mortality [19, 28, 29]. In addition, light PA has beneficial effects on the health situation of elderly people, e.g. a higher cognitive performance [30], and it is positively associated with physical health and life satisfaction [31]. Hence, the impact of light PI on public health may be underestimated to some extent.

Increasing age is associated with decreasing regular PA, a decreasing number of daily steps [32] and an increase of PI which contributes to higher risks for chronic diseases (e.g. stroke) and multimorbidity [10, 33–36]. In the context of demographic change, particularly older people should be an important target group for extensive and practicable prevention strategies.

Interventions in the elderly have the potential to increase PA [37-39] and to reduce PI [40-42]. Several pedometer-based intervention studies have shown positive effects on the level of PA. Harris et al. showed a positive effect of an intervention consisting of four consultations by a primary care nurse on daily steps and weekly MVPA in ≥10 min bouts.

A meta-analysis on pedometer-based PA interventions pointed out that usage of a pedometer has a moderate positive effect on PA levels (average increase of 2,000 daily steps) [38]. These results correspondent to a systematic literature review in which participants in the intervention group increased their PA by 2,491 daily steps by using a pedometer-based intervention compared to the control group [39].

Interventions also have the potential to decrease PI [43]. Mutrie et al. have shown that a nurse-delivered intervention in the elderly (N = 41), based on social cognitive theory, reduced daily PI by 68 minutes in the intervention group (measured by pedometer and accelerometer) compared to the controls [44]. Another study of older adults (N = 478) showed that a home-based intervention based on social cognitive theory led to a decrease of 57 minutes of sitting time per day in the intervention group compared to the control group [42].

Most of the successful studies are based on complex interventions (i.e. personal consultations) [37]. These are associated with high costs and a large demand of personal resources. Dissemination of such interventions to the general older population may be hardly feasible, especially against the background of the demographic change. In conclusion, a translation of such interventions into daily life of a major proportion of the elderly has been difficult. Therefore, we designed a low-threshold and practicable intervention to increase overall PA and decrease PI.

Research objectives

Aim of the study was the evaluation of a low-threshold intervention to improve overall PA and simultaneously reduce PI. The objective of this analysis was to examine the effect of a low-threshold intervention, consisting of automatically generated, individualized feedback-letters reporting the measured physical activity on PA and PI in a sample of apparently healthy elderly people in the Northeast of Germany (MOVING-study (Motivation-Oriented intervention study for the elderly in Greifswald). Primary outcomes of the analysis were overall PA and PI after 6 months in the intervention group compared to the control group.

Materials and methods

Study sample

A detailed study protocol [45], as well as levels and determinants for PA and PI of the study sample of the MOVING study at baseline are published elsewhere [46]. In brief, the study population consists of people sampled from the general population who meet the following inclusion criteria:

Age ≥ 65 years

The possibility of being physically active in daily life

Exclusion criteria were:

Permanent use of a wheelchair (no ability to walk independently)

Simultaneous participation in other studies addressing PA and PI

Not accessible by telephone or cell phone

Fulfillment of the WHO recommendations for PA (self-report) for people aged ≥ 65 at baseline (≥150 minutes of moderate PA or 75 minutes of vigorous PA)

All eligible patients were informed in detail about the study and had to give their written informed consent prior to study inclusion. The project data was assessed using eCRFs in an IT-supported documentation system [47]. At Baseline, all participants received a study ID-card with individual 1D-barcode for reliable re-identification.

Study procedure

Potential participants and interested persons contacted the study staff by telephone. A short telephone interview was performed to check the inclusion and exclusion criteria. All persons who fulfilled the inclusion criteria were invited to the examination center of the German Centre for Cardiovascular Research (DZHK) of the University Medicine Greifswald. Here, all participants received a 3-axis accelerometer (ActiGraph GT3x-BT, Pensacola, FL, USA) and specific instructions on how to wear the accelerometer device.

After the baseline examination, participants were individually randomized 1:1 into the control and the intervention group. Follow-up examinations were conducted in both groups at 3 and 6 months after the baseline examination with the same program as in the baseline examination.

Measures

All study participants received a physical examination consisting of blood pressure measurement and measurement of body weight, waist and hip circumference. Trained and certified study staff carried out all measurements according to standard operating procedures (SOP). At baseline, participants were asked for sociodemographic data (sex, age, education (number of years participants went to school)). In addition, participants received standardized paper-pencil questionnaires on PA and PI (IPAQ [48], German PAQ 50+ [49], MOST [50], SF8 [51] and SSA-scale [52]) that had to be completed after the 7-days of wearing the accelerometer device.

After the baseline and at the two follow-up examinations, all study participants wore the accelerometer device at daytime on the right hip. The participants were instructed to remove the device only for water-based activities. The three-axis accelerometer recorded PA, PI and sedentary breaks (a period of non-sedentary activity (e.g. walking) in between two sedentary conditions (e.g. sitting) continuously at a sampling frequency of 30 Hz over a period of seven consecutive days, starting at midnight after the examination day.

The software ActiLife 6.13.2–6.13.3 (ActiGraph Cop. ©, Pensacola, FL, USA) was used for data download and data processing. Raw data were calculated into 60-second epochs and saved as GT3X files.

We used specific cut points based on Freedson [53] to categorize the intensity of participants’ PA. For adults, this cut-point model is considered as established for the determination of different physical activity categories [54]. Intensity of PA is categorized as sedentary (0–99 counts), light (100–1951 counts), moderate (1952–5724 counts), and vigorous (5725–9498 counts) PA. Non-wear time was defined as ≥ 60 min of continual zero counts.

Intervention

The intervention based on individual accelerometer data and consisted of automatically generated, personalized feedback letters sent by mail. The participants in the intervention group received the letters after the baseline examination and the 3-month follow-up shortly after wearing the accelerometer. The participants in the control group received no feedback during the study period. According to the Declaration of Helsinki, we provided participants in the control group a detailed feedback about PA and PI after the end of the study within a thank you letter. This procedure ensured that all participants (both groups) received in sum the same information at the end of the study.

The feedback letters contained three graphs, reporting on the measured PA and PI, with provision of the number of steps per day, daily and weekly moderate and vigorous PA in minutes and daily and weekly sedentary time in minutes, as well as a paragraph about the importance of sedentary breaks. The participants received age appropriate recommendations for PA based on the WHO-guidelines [26]. The feedback letters were automatically generated in R software (version 3.3.2, Lucent Technologies, Murray Hill, NJ, USA).

Data analysis

To examine the effects of the intervention, we analyzed overall, light and moderate PA as well as PI as dependent variables with a Two-Way Mixed ANOVA with repeated measures with group (between subject) and time (inner subject) as factors. For baseline data, we analyzed the equality of variables between both study groups. Due to significant difference between both groups in partnership (p = 0.006), we included partnership as a covariable in the Two-Way Mixed ANOVA. First, we analyzed the assumptions and requirements of mixed ANOVA i.e. distribution and homoscedasticity and detected no violation of the assumptions”. Bonferroni adjustment was made because several analyses were performed.

All statistical analyses were performed using pseudonymized data. In addition, we performed a sub-group analysis and excluded participants with ≥ 150 minutes of weekly moderate PA or ≥ 75 minutes of weekly vigorous PA at baseline (according to the WHO recommendation).

A valid measurement day was defined as a record of at least 10 hours of total wearing time during the day. Additionally, a record time of at least 4 valid days was required for all data analyses.

Statistical significance was assumed for p-values <0.05. Data processing and statistical calculations were performed with IBM SPSS Statistics 26 or later (1989–2020 by IBM Corp. ©, Armonk, New York, USA.).

Ethical consideration

In 2016, the clinical ethics committee of the University Medicine Greifswald (protocol number BB071/16) approved the study.

Results

A number of 258 potential participants were contacted personally or by telephone, thereof 206 were included in the study and randomized (Fig 1).

Fig 1

Consort diagram: Recruited participants and participants included in the analysis (N = 166).

At baseline, the participants had a mean wearing time of the accelerometer of 5,934.5 minutes (SD = 789.5) per week, which corresponds to 14.1 hours daily. N = 166 participants could be included in the analysis, thereof n = 97 women (58.4%) and n = 69 men (41.6%). At baseline, the mean age was 70.8 years (SD = 4.8) (Table 1).

Table 1

Descriptive characteristics of the study sample (N = 166), complete and valid data sets (baseline, 3 and 6 months follow-up).

	Control group		Intervention group		Both groups
Characteristics	n	Mean (SD) or n (%)	n	Mean (SD) or n (%)	n	Mean (SD) or n (%)
Sex (women)	81	47 (58.0%)	85	50 (58.8%)	166	97 (58.4%)
Age (yr)	81	71.2 (SD 5.0)	85	70.4 (SD 4.6)	166	70.8 (SD 4.8)
Number of participants currently living in a partnership (yes)*	77	47 (61.0%)	83	67 (80.7%)	160	114 (71.3%)
Education (yr)	74		82		156
< 10 years		14 (18.9%)		18 (22.0%)		32 (20.5%)
= 10 years		25 (33.8%)		17 (20.7%)		42 (26.9%)
> 10 years		30 (40.5%)		45 (54.9%)		75 (48.1%)
Other		5 (6.8%)		2 (2.4%)		7 (4.5%)
Body mass index (kg/m2)	81		85		166
< 25		22 (27.2%)		23 (27.1%)		45 (27.1%)
≥ 25 and < 30		33 (40.7%)		30 (35.3%)		63 (38.0%)
≥ 30		26 (32.1%)		32 (37.6%)		58 (34.9%)
Waist circumference (cm)	81	96.7 (SD 13.1)	85	94.8 (SD 14.6)	166	95.8 (SD 13.9)
Hip circumference (cm)	81	104.1 (SD 11.8)	85	102.7 (SD 10.4)	166	103.4 (SD 11.1)
Blood pressure (sys/dia mmHg) (Pulse /min)	81	133/74 (70)	84	132/72 (73)	165	133/73 (71)
Wearing time of the accelerometer
Baseline	81	5,922.7 (SD 926.8)	85	5,945.7 (SD 637.2)	166	5,934.5 (SD 789.5)
3-month follow-up	81	5,833.6 (SD 839.1)	84	5,826.4 (SD 683.9)	165	5,829.9 (SD 761.7)
6-month follow-up	79	5,636.4 (SD 719.7)	82	5,840.7 (SD 719.0)	161	5,740.5 (SD 724.4)

Notes: n number of subjects, SD standard deviation *significant difference between intervention and control group (p = 0.006).

At baseline, the average time of overall PA was 2473.7 (95% CI 2387.0–2560.4) minutes per week, to which light PA contributed the largest part with 2260.9 (95% CI 2181.1–2340.7) minutes. In addition, the average of weekly PI (baseline) was 3460.8 (95%-CI 3337.5–3584.0) minutes per week which correspondents to about 8.2 hours per day and about 57% of participants’ waking time.

Participants in the intervention group slightly increased their weekly time of light PA from baseline to 6-months follow-up on average by + 54.0 minutes (control group—59.2 minutes). The average overall PA increased in the intervention group by 25.0 minutes per week (control group—75.9 minutes) between baseline and 6 months follow-up. At 6-months follow-up (primary endpoint), the weekly difference of overall PA between intervention group (2488.8 minutes, 95% CI 2358.9–2618.2) and control group (2408.2 minutes, 95% CI 2263.0–2553.4) was 80.6 minutes which correspondents to about 1.3 hours of additional overall PA per week in the intervention group (Table 2).

Table 2

Physical activity and inactivity (mean number of minutes during the wearing-week), by intensity of physical activity and overall PA (N = 166).

Study phase	Study Group	Physical inactivity	Sedentary breaks	Light PA	Moderate PA	Vigorous PA	Overall PA	Steps Counts
		Mean (CI 95%)	Mean (CI 95%)	Mean (CI 95%)	Mean (CI 95%)	Mean (CI 95%)	Mean (CI 95%)	Mean (CI 95%)
Baseline	Intervention	3482.0	97.6	2246.4	213.6	3.8	2463.8	99,278.0
	group (n = 85)	(3336.1–3627.8)	(92.7–102.4)	(2139.9–2352.8)	(180.6–246.6)	(0.6–7.0)	(2342.8–2584.8)	(92,889.8–105,666.2)
	Control	3438.5	95.9	2276.1	201.1	6.6	2484.1	97,763.2
	group (n = 81)	(3234.3–3642.7)	(89.3–101.9)	(2154.4–2397.8)	(166.5–235.6)	(1.2–14.3)	(2357.2–2611.1)	(90,840.4–104,685.9)
	Total (n = 166)	3460.8	96.8	2260.9	207.5	5.1	2473.7	98,538.8
		(3337.5–3584.0)	(93.0–100.6)	(2181.1–2340.7)	(183.8–231.1)	(1.1–9.2)	(2387.0–2560.4)	(93,884.7–103,193.0)
3-months follow-up	Intervention	3351.1	92.4	2292.3	180.3	2.7	2475.3	99,554.3
	group (n = 84)	(3204.8–3497.3)	(87.6–97.0)	(2174.7–2409.8)	(148.3–212.3)	(0.2–5.6)	(2348.7–2602.0)	(93,268.3–105,840.2)
	Control	3318.8	93.6	2327.5	181.5	5.3	2514.8	97,567.4
	group (n = 81)	(3140.0–3497.7)	(87.8–99.4)	(2188.9–2466.2)	(151.0–212.0)	(1.0–9.7)	(2369.9–2659.6)	(90,803.8–104,330.9)
	Total (n = 165)	3335.3	93.0	2309.6	180.9	4.0	2494.7	98,578.9
		(3221.3–3449.2)	(89.4–96.6)	(2219.9-	(159.0–202.8)	(1.4–6.6)	(2399.7–2589.7)	(94,013.7–103,144.1)
6-months follow-up	Intervention	3354.8	92.5	2300.4	184.9	3.3	2488.8	99,613.6
	group (n = 83)	(3210.0–3499.6)	(88.1–96.9)	(2180.1–2420.8)	(154.2–215.6)	(0.7–5.8)	(2358.9–2618.2)	(93,526.0–105,701.2)
	Control	3228.2	90.2	2216.9	187.7	3.5	2408.2	94,301.4
	group (n = 79)	(3087.0–3369.4)	(85.6–94.7)	(2077.8–2356.0)	(150.3–225.1)	(1.6–8.5)	(2263.0–2553.4)	(87,290.4–101,312.5)
	Total (n = 162)	3293.1	91.4	2259.7	186.3	3.4	2449.5	97,023.1
		(3192.5–3393.7)	(88.2–94.5)	(2168.8–2350.6)	(162.5–210.1)	(0.6–6.2)	(2353.2–2545.8)	(92,426.6–101,619.7)

Notes: n number of subjects, CI 95% confidence interval.

At baseline, both study groups started almost on the same value of overall PA. At 6-months follow-up, the amount of overall PA decreased in the control group (red line) the baseline value while participants in the intervention group (blue line) could slightly increase their weekly overall PA (Fig 2).

Fig 2

Mean minutes of overall PA (95% CI, intervention group and control group) at baseline, 3 and 6 months follow-up (N = 166), control group (red line), intervention group (blue line).

At baseline, n = 97 (58.4%) of the study participants (both study groups) achieved the international recommendations for moderate PA (≥150 minutes moderate or ≥ 75 minutes vigorous PA). The proportion of participants who fulfilled the recommendations slightly decreased over study time (Table 3).

Table 3

Number and proportion of study participants who fulfilled the WHO recommendations PA for people aged over 65 years, separately for intervention and control group (N = 166).

WHO recommendation	Study group	Baseline (n = 166, IG = 85, CG = 81)	3-month follow-up (n = 165, IG = 84, CG = 81)	6-month follow-up (n = 162, IG = 83, CG = 79)
moderate PA ≥150 min. per week or	Intervention group	49 (57.6%)	43 (51.2%)	41 (49.4%)
vigorous PA ≥75 min. per week	Control group	48 (59.3%)	39 (48.1%)	38 (48.1%)
	Total	97 (58.4%)	82 (49.7%)	79 (48.8%)

n number of subjects, fulfillment of WHO recommendations for moderate PA for people aged 65 and older: ≥150 minutes moderate PA or ≥75 minutes vigorous PA or an equivalent combination per week (2 minutes moderate PA correspond to 1 minute vigorous PA).

There was no statistically significant interaction between time and group (interaction effect: time*group) for the depending variables PI, light, moderate and overall PA (sphericity assumed). P-values were ranging between 0.317 to 0.546. Physical inactivity (n = 146)

The results of the Two-Way Mixed ANOVA showed that there was a significant main effect of group, F(1, 143) = 5.335, p = 0.022, partial η2 = .036. In addition, there was no significant main effect of time, sphericity assumed F(2, 286) = 0.763, p = 0.467, partial η2 = .005.

Descriptive statistics showed that both study groups decreased the amount of PI over study time. At 6-months follow-up, participants included in the analyses had an amount of 3310.4 minutes (SD = 536.1) of PI (mean) in the intervention group, and 3161.1 minutes (SD = 581.8) of PI (mean) in the controls. Besides that, the differences between both groups were about the same amount of minutes in PI at each measurement point (baseline, 3- and 6-months follow-up).

Light PA (n = 155)

The results of the Two-Way Mixed ANOVA showed that there was no significant main effect of group, F(1, 153) = 0.017, p = 0.895, partial η2 = .000, and time, sphericity assumed F(2, 304) = 2.196, p = 0.113, partial η2 = .014.

Descriptive statistics showed almost the same baseline value for light PA between the intervention (2235.8 minutes (SD = 495.7) and control group (2232.1 minutes, SD = 536.4). After an increase of light PA in both study groups to 3-months follow-up, participants in the control group decreased minutes of light PA even lower than the baseline value (2187.4 minutes, SD = 604.5) at 6-months follow-up. Besides that, participants in the intervention group remained almost at the same amount of light PA at 6-months follow-up (2298.8 minutes, SD = 557.8). Thus, participants in the intervention group showed a higher amount of light PA at 6-months follow-up (primary endpoint) compared to the controls (+111.4 minutes of light PA) in the descriptive statistics.

Moderate PA (n = 147)

The results of the Two-Way Mixed ANOVA showed that there was no significant main effect of group, F(1, 144) = 1.426, p = 0.234, partial η2 = .010, and time, sphericity assumed F(2, 288) = 0.715, p = 0.490, partial η2 = .005.

Descriptive statistics showed a constant decline of moderate PA in the control intervention group starting from baseline (172.4 minutes, SD = 124.1) to 6-month follow-up (145.9 minutes, SD = 115.1). Participants in the intervention group started with an average amount of moderate PA about 201.8 minutes (SD = 149.6) and could increase the amount after a decline at 3-months follow-up (163.9 minutes, SD = 127.4) at 6-months follow-up to 170.9 minutes (SD = 126.3).

Overall PA (n = 155)

The results of the Two-Way Mixed ANOVA showed that there was no significant main effect of group, F(1, 152) = 0.005, p = 0.945, partial η2 = .000, and time, sphericity assumed F(2, 304) = 1.638, p = 0.196, partial η2 = .011.

Descriptive statistics showed slightly increase (+36.1 minutes) of the amount of overall PA in the intervention group starting at 2459.6 minutes (SD = 558) to 3-months follow-up (2495.7, SD = 564.4). Participants in the intervention group stayed almost on this level at 6-months follow-up (2494.6, SD = 590.8). The controls started almost on the same baseline value (2459.0, SD = 581.1). After an increase to 3-month follow-up (2504.7, SD = 657.2), the amount of overall PA declined to 6-months follow-up below the intervention group (2399.4, SD = 661.4). At 6-months follow-up, the group difference was 95.2 minutes of weekly overall PA.

Sensitivity analyses

In a sensitivity analysis, we excluded participants with ≥ 150 minutes of weekly moderate PA or with ≥ 75 minutes of vigorous PA at baseline (according to the WHO recommendation). Here (n = 153), we found a statistically significant interaction effect for time*partner (covariable) for light PA, sphericity assumed F(2, 300) = 3.020, p = 0.05, partial η2 = 0.020. This result indicate a small positive effect on light PA over study time in participants who are in a partnership.

Discussion

The results of this analysis show that the levels of light, moderate and overall PA in our sample were exceptionally high for this age group [55-57]. Both groups started on almost the same values of overall PA at baseline (intervention group = 2463.8 (95% CI 2342.8–2584.8) minutes, control group = 2484.1 (95% CI 2357.2–2611.1) minutes. The level of overall PA over both groups remained almost on the same level at the 3-months follow-up (intervention group = 2475.3 (95% CI 2348.7–2602.0) minutes, control group = 2514.8 (95% CI 2369.9–2659.6) minutes), and 6-months follow-up (intervention group = 2488.8 (95% CI 2358.9–2618.2) minutes, control group = 2408.2 (95% CI2263.0–2553.4) minutes). At 6-months follow-up, participants in the intervention group showed more overall PA compared to the participants in the control group (difference = 80.6 minutes per week which correspondents to about 1.3 hours of additional overall PA per week in the intervention group).

In previous randomized controlled intervention trials promoting PA increasing PA levels have often been observed also in the control groups [58, 59]. Therefore, several possible reasons have been proposed, e.g. the “Hawthorne effect”, which may have changed participants’ behavior as a result of participating in the study–irrespective of their randomisation. In our study, we observed this effect according to the amount of PI, both study groups (including controls) decreased PI over study time.

However, in a longitudinal analysis at the 6-month follow-up, the levels of light and overall PA had decreased in the control group (below the baseline value), while PA in the intervention group remained almost at the same level. Seasonal influences, (weather) cannot be the cause for the difference, because the intervention and the control group were simultaneously recruited and included between November 2016 and December 2018.

Also the amount of PI was high and with 3460.8 (95% CI 3337.5–3584.0) minutes per week at baseline, which corresponds to 8.2 hours/day and is comparable with other studies examining the same age group [14, 60]. Thus, participants in both study groups spent on average about 57% of their waking time in PI at baseline. The high amount of PI combined with a parallel high amount of PA indicates that PI and PA can be seen as independent factors.

The intervention showed no significant effect regarding the primary outcomes. However, a large part of the participants was already physically active on a high level: 58.4% (n = 97) of the participants already met the recommendations for people aged over 65 years of the WHO at baseline. This limits the potential to increase the participant’s PA. Sensitivity analyses excluding very active participants, showed significant effects of the intervention on overall PA among the subgroup of the less active. Thus, further research should focus more on people with low PA levels. However, as a consequence of some self-selection of the more physically active elderly, recruitment of people with a low level of PA may be challenging.

The layout of interventions to increase PA and reduce PI is crucial. As previous research has shown, interventions should focus on the target group and especially on individual factors (e.g. age, sex) [46]. In addition, an inclusion of environmental and interpersonal factors can be helpful to change participants’ behavior [46].

The study participants showed a good adherence to wearing the accelerometer. At baseline, the mean wearing time was on average 5934.5 minutes (SD = 789.5), which correspondents to 14.1 hours per day. Only n = 15 participants had to be excluded from the analysis because of too little wearing time over the six months follow-up time. In conclusion, our data support that the use of an accelerometer is feasible and practical in this age group.

Previous research has shown that intervention studies have the potential to increase PA and decrease PI [37-42]. Most of the previous studies are characterized by rather complex and time-consuming interventions, e.g. personal consultations or telephone calls, which limits the transfer of successful concepts to the general population. Our intervention used individualized feedback-letters based on measured PA and PI and was developed to examine a low-threshold intervention with high practicability and a high potential for translation into the general population. Due to some ceiling effect, the overall intervention effect reduced somewhat.

Especially elderly people are a relevant target group for promoting light PA, because an increase of moderate and vigorous PA might be harder to achieve for this age group [61]. Presently, quantitative recommendations towards PI are lacking and the public health message comes down to “move more at any intensity” [6, 15]. Thus, light PA should be included in global recommendations [62].

Several studies have shown that effective interventions to increase PA consider behavioral-change aspects and address participants’ needs and beliefs. Thus, targeting and tailoring are two important aspects of successful interventions [63].

The number of intervention studies promoting PA and reducing PI increased continuously over the last decade. While there is international consensus regarding PA recommendations, it is too early to implement quantified recommendations regarding PI, although some countries (e.g. Australia) introduced general guidelines to reduce sitting time in general and break up long periods of sitting regularly [25]. In addition, there is an ongoing discussion, whether PI is an independent risk factor for mortality [64], and there is still no global unified definition for PI. Thus, further research regarding PI and specific and quantitative recommendation are urgently needed.

Limitations and strengths

We recruited study participants in a convenience sample using a variety of recruitment methods. Some of these methods have likely increased the proportion of participants with above average PA compared to PA in the general population–so the composition of our study sample somewhat limits the generalizability of our results. Due to this fact, a selection bias can not be excluded. Prior to recruitment, we tried to exclude very active people from the study (based on self-assessment) to address a possible ceiling effect. As the results have shown, many participants underestimate their own amount of PA. Therefore, we calculated a sensitivity analysis excluding very active participants with an amount of ≥150 minutes of moderate PA. Some ceiling effect reduced the overall intervention effect somewhat.

In general, accelerometry is an internationally established method for the objective assessment of physical activity and allows a valid and reliable record of PA intensity, frequency, and duration [54, 65]. Although, currently no single specific device can be recommended for all purposes, since the choice of the device depends on the research question, design and target group [54]. However, data from the accelerometer device can potentially differ from the real levels of PA and PI especially in the elderly, because several activities are carried out in standing (i.e. gardening) or sedentary positions (i.e. gymnastic on stools) which cannot be assessed accurately. Only few missings were obtained for the dependent variables (n = 4).

Conclusion

Global demographic changes are associated with a large increase of the number of older people and a simultaneous growth of non-communicable-chronic diseases (i.e. cardiovascular disease). Therefore, there is a high need for effective prevention strategies. Although we detected no significant differences in the primary outcome in the interaction effect (time*group) between the study groups, descriptive statistics give a hint of a small positive effect of the intervention. Sensitivity analysis has shown, that partnership is a significant positive factor for light PA (time*partnership). Thus, further research should focus on the recruitment of the adequate target groups (especially people without partner and low levels of PA and high levels of PI). Therefore, local stakeholders should be consulted and setting related factors should be considered.

Most of the successful studies are based on complex and/or resource consuming interventions (i.e. personal consultations), which is accompanied by high costs and a large amount of personal resources. In the future, the usage of internet-based communications and e-health applications (e.g. video-consultation) can support the provision of information and interventions, which makes a translation of (complex) interventions into daily life easier.

Results of our study confirm that a low-threshold intervention with low costs and little amount of personal resources can be implemented and is able to hold up adherence of elderly people. Our analyses shows limited and not significant, but seems to have a small positive effect of a low-threshold intervention after 6 months follow-up.

Notes: n number of subjects, SD standard deviation.

(PDF)

Click here for additional data file.

Notes: n number of subjects, CI 95% confidence interval, M mean.

(PDF)

Click here for additional data file.

(PDF)

Click here for additional data file.

Dataset.

(XLSX)

Click here for additional data file.

17 Mar 2021

PONE-D-21-04633

A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study

PLOS ONE

Dear Dr. Kleinke,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Even though some reviewers consider that major revision is required and others consider that the manuscript should be rejected, I have decided not to reject it due to the relevance of the problem under study.

In the Introduction, there is redundant information that could be condensed. Is it necessary to talk about PA and PI? My question is because if they are opposites, only one of these terms should be used.

The objective of the treatment (increasing PA) must be differentiated from the objective of the study (exploring the effectiveness of the treatment).

In the Methods section, it does not seem to be clear how the sample size is N = 157; even one reviewer comments that the initially defined inclusion criteria are being violated.

The authors say: "To examine the effects of the intervention, we analyzed the overall, light, moderate and vigorous PA and PI after 6 months in the intervention group compared to the control group". These qualitative levels of PA have not previously been defined.

The difference between intervention and non-intervention should be made explicit. Was it only that the experimental group received the letters and the control group did not receive them? According to the Declaration of Helsinki regarding placebo use, what was done with the control group at the end of the study?

Please specify what the variables "education" and "health status" measure.

I agree with the reviewers that statistical analysis is not appropriate. Reviewer 2 suggests applying ANOVAs and considers it of interest to explore the effect of other variables (age, sex, etc.). Please report differences in age, sex, years of education, BMI, waist and hip circumference, blood pressure, etc., between groups; maybe it will be necessary to include some of these variables as covariables to mitigate their effects.

I also agree with Reviewer 2 concerning some conclusions that are not supported by the results.

Please submit your revised manuscript by April 17. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Thalia Fernandez, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Thank you for stating the following financial disclosure:

"The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript."

At this time, please address the following queries:

Please clarify the sources of funding (financial or material support) for your study. List the grants or organizations that supported your study, including funding received from your institution.

State what role the funders took in the study. If the funders had no role in your study, please state: “The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

If any authors received a salary from any of your funders, please state which authors and which funders.

If you did not receive any funding for this study, please state: “The authors received no specific funding for this work.”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: No

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I will focus on methods and discussion

Major

1) As the authors acknowledge this is a select population that makes generalisability of the findings questionable.

2) Some information on the randomisation procedure and how it was implemented is needed.

3) The sample size has not been arrived to through a power calculation but it is a convenience sample.

4) Results are poorly reported. Specifically the authors needs to make clear:

a) Drop outs - from 206 to 157?

b) report characteristics by group (i.e. Table 1). was balance across these key parameters achieved?

c) if not (i.e. b) then a t-test is not the apprpriate analysis

e) a secondary outcome should have been time wearing the device and that should have been reported in the abstract as well

f) report drop outs by group. Also the number of drop outs is relatively large for a keen sample of healthy adults.

5) No mention of missing data. was multiple imputation considered to boost the sample and get more robust estimates, compared to a complete case analysis?

Minor

1) Abstract, discussion: no clear mention of the intervention and why it may not work.

2) A more deterministic approach could have been used to assign people to groups, something often used when samples are modest, as they are here (to ensure good balance on key characteristics).

Reviewer #2: Authors evaluated the effectiveness of a low-threshold intervention to increase PA in elderly. Two groups (intervention-control) were recruited. No differences in most of the variables were found. Authors concluded that PA could be improved in participants with low levels of PA in both groups of this low-threshold intervention.

Several concerns regarding the rationale, methodology, results and discussion are explained below.

Authors propose “targeting and tailoring are two important aspects of successful interventions” on basis of the participants’ needs and beliefs, however, what they propose here “a low-threshold intervention” does not meet this criterion. Please justify and show evidence (if the case) of the effectiveness of a more “practicable intervention”, maybe from studies were only biofeedback is used to increase PA.

Table 1. Why the number of participants (n) change across the descriptive characteristics, for instance, there are more (161) participants in the line “Number of participants

currently living in a partnership (yes)” than those included in the study.

Please, include a Table showing the “descriptive characteristics” between groups and the statistical comparisons to exclude the presence of significant differences in these variables.

Explain, in the exclusion criteria the cutoff for the fulfillment of the PA according to WHO that you considered. In this regard, if this was an exclusion criterion, why the whole sample was included in the first analysis, and just excluded for the subsequent subgroup analysis?

Explain, in data analysis, how delta values were calculated.

At methodological level, I see two main problems, 1) participants with ≥ 150 minutes of weekly moderate PA at baseline (according to the WHO recommendation) which actually fulfillment an exclusion criterion, were included. 2) The statistical approaching is not justified for this data, why to use t-test instead of two way anova with group (2) and follow-up (3) as factors? The latter could help you to understand within subjects and between groups differences. On the other hand, why data of the first follow-up is not analyzed but included in Tables and Figures?

The only significant result was a difference in delta value between groups, however, this result did not confirm a PA increase due the intervention, but a decrease in the PA in the control group, according with Figure 2, is that right? Therefore, the effectiveness of the intervention is not corroborated?

In the first paragraph of the discussion, authors stated that the intervention group showed more overall PA compared to the participants in the control group, however this result was not significant. Therefore, no differences between groups were observed at any time.

In lines 286-287 authors suggest “Interventions should focus on the target group and especially on individual factors (e.g. age, sex)”, but not analysis was carried out with these variables in this study, it could be an interesting approaching. It is known, in previous literature, the differences between sexes in terms of PA and self-efficacy to PA.

From lines 301-313 authors discuss the importance of light PA increase and PI decrease, but their results did not show evidence of the effectiveness of the intervention even for these levels of PA.

In the conclusion authors suggest “Low-threshold interventions have the potential to increase PA and reduce PI in the general elderly population” however their results did not support this statement.

Reviewer #3: The paper requires a new organization of the introduction section. It is hard to read. Additionally, the statistical analysis and the results reported seem to be insufficient for the study's aims. I suggest major revision based on these aspects.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes: Graciela Catalina Alatorre-Cruz

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: R-1.docx

Click here for additional data file.

29 Apr 2021

PONE-D-21-04633

A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study

PLOS ONE

Response to Reviewers

Thank you for coordinating the review process and for your constructive comments to the manuscript. In our opinion, the comments help to increase the quality of the manuscript, especially the part of the statistical analyses. We have revised the manuscript and addressed all reviewer’s comments.

Comment: In the Introduction, there is redundant information that could be condensed. Is it necessary to talk about PA and PI? My question is because if they are opposites, only one of these terms should be used.

Response: Thank you for the comment. To our knowledge, “we still know little about the independent detrimental health effects of sitting, and the possibility that sitting is mostly the inverse of physical activity remains”. (Stamatakis, 2018a). In addition, the evidence about sedentary behavior is “underdeveloped and inconsistent; it cannot support quantitative guidance” yet.

Literature points out, that PA is not “only” the opposite of PI. Rather, PA and PI are in an interplay that has not yet been finally clarified (Ekelund, 2018, Stamatakis, 2018b). In our opinion, it is therefore important to analyze both factors and to mention them in the introduction. In sum, “Understanding the synergistic and independent effects of (ideally objectively measured) MVPA, standing, light intensity PA and sitting on CVD risk is a future research priority” (Stamatakis, 2018b). For this reasons, it is necessary to talk about both, PA and PI.

Comment: The objective of the treatment (increasing PA) must be differentiated from the objective of the study (exploring the effectiveness of the treatment).

Response: Thank you for your comment. In the revised version of the manuscript, we have addressed your comment and emphasized the difference between the treatment and the objective of the study more clearly at the corresponding points in the manuscript. The objective of the study is the evaluation of a low-threshold intervention to improve PA and reduce PI among older study participants. The objective of the treatment is to improve PA and reduce PI. (please see lines 135 - 136)

Comment: In the Methods section, it does not seem to be clear how the sample size is N = 157; even one reviewer comments that the initially defined inclusion criteria are being violated.

Response: Thank you for this comment. For the traceability of the dropouts for each study group, we provide a Consort diagram (Fig. 1) according to PLOS ONE requirements. The Consort diagram is an attached file (supporting information) – please see Fig. 1 “S1 Fig. Consort diagram: Recruited participants and participants included in the analysis (n = 166)”.

After a carefully check of the data, we could increase the number of complete data sets about n = 5. Due to Intention-to-treat principle, we also included n = 4 incomplete datasets. In sum, we could increase the study sample about n = 9, so the study sample in the revised manuscript is now n = 166 at baseline. Due to the small number of incomplete data sets at Follow-up 1 (n=1) and Follow-up 2 (n=3), we decided us against an imputation of data. For the main analysis (Two-Way Mixed ANOVA), the number of data sets included differs depending on the variable, because of excluded outliers.

The exclusion criteria of the study was the fulfillment of the WHO recommendations for PA for people aged ≥ 65 at baseline based on self-report and not by using objectively accelerometer data. Nevertheless, some participants evidently underestimated their own weekly amount of PA before participating in the study. When wearing the accelerometer, the amount of PA could be quantified. A subsequently exclusion of the very active participants (≥150 minutes moderate PA | ≥75 minutes vigorous PA) appeared to us to be inappropriate from a methodological point of view.

We defined all inclusion and exclusion criteria of the study a priori. The self-assessment was about the degree of time of weekly moderate physical activity was included to address a possible ceiling effect of PA in this age group. The operationalization of this exclusion criteria focused on self-reported degree of weekly moderate PA prior to recruitment, and therefore independent of the outcomes of the study. Thus, study participants whose self-reported PA exceeded the WHO recommendation of ≥ 150 minutes of moderate PA (objectively measured with an accelerometer device), were not excluded from the main analysis and only used for a sub-group analysis.

Comment: The authors say: "To examine the effects of the intervention, we analyzed the overall, light, moderate and vigorous PA and PI after 6 months in the intervention group compared to the control group". These qualitative levels of PA have not previously been defined.

Response: In response to your concerns about the qualitative levels of PA, we moved the following sentences in the measures section and added explicit information to the manuscript: “We used specific cut points based on Freedson to categorize the intensity of participants’ PA. For adults, this cut-point model is considered as established for the determination of different physical activity categories. Intensity of PA is categorized as sedentary (0–99 counts), light (100–1951 counts), moderate (1952–5724 counts), and vigorous (5725–9498 counts) PA. Non-wear time was defined as ≥ 60 min of continual zero counts.” (please see lines 188 – 192)

Comment: The difference between intervention and non-intervention should be made explicit. Was it only that the experimental group received the letters and the control group did not receive them? According to the Declaration of Helsinki regarding placebo use, what was done with the control group at the end of the study?

Response: Thank you for this comment. The focus of our study was to examine the effect of a low threshold intervention. Only study participants in the intervention group received two individual feedback letters containing objectively measured PA and PI times based on data from the accelerometer device captured at baseline and 3 months follow-up measurement. Thus, this was the difference between the experimental (intervention) and control group.

Participants in the control group did not receive any feedback while study time. According to the Declaration of Helsinki, we provided participants in the control group a detailed feedback about PA and PI after the end of the study within a thank you letter. This procedure ensured that all participants (both groups) received in sum the same information at the end of the study. In conclusion, there were no differences in type, access and scope of information between both groups (intervention and control).

In response to your concerns about the explicit difference between the intervention and control group, we added the following sentence to the manuscript in the intervention section: “The participants in the control group received no feedback during the study period. According to the Declaration of Helsinki, we provided participants in the control group a detailed feedback about PA and PI after the end of the study within a thank you letter. This procedure ensured that all participants (both groups) received in sum the same information at the end of the study.” (please see lines 197 – 201)

Comment: Please specify what the variables "education" and "health status" measure.

Response: The education variable measured the number of years participants went to school (< 10 years, = 10 years, > 10 years). To address your concern, we specified the variable in the revised version of the manuscript and added the following sentence to the measures section: “number of years participants went to school”. (please see line 175)

In this manuscript, we did not include health status as a variable or analyzed it separately. For this reasons, we excluded the term “health status” from the manuscript. We now provide a more streamlined version of the measures section. Thank you for this comment.

Comment: I agree with the reviewers that statistical analysis is not appropriate. Reviewer 2 suggests applying ANOVAs and considers it of interest to explore the effect of other variables (age, sex, etc.). Please report differences in age, sex, years of education, BMI, waist and hip circumference, blood pressure, etc., between groups; maybe it will be necessary to include some of these variables as covariables to mitigate their effects.

Response: Thank you for this comment to the statistical analysis. In the revised version of the manuscript, we provided a new statistical analysis: Two-Way Mixed ANOVA with repeated measures that is in line with reviewer 2. To address your concerns about the effect of other variables, we added the variables waist, hip circumference, blood pressure and pulse in the table 1 (descriptive characteristics of the study sample) that is consistent with Reviewer 2 recommendations (please see page 9/10). In addition, we checked all variables regarding significant group differences. Here, we detected a systematic difference in the variable “partnership” between intervention and control group. Therefore, we added the variable “partnership” as a covariate in the Two-Way Mixed ANOVA to mitigate the effect. For the ANOVA, we checked all assumptions and requirements (please see details below).

In the revision, we added the new data file including the added variables (waist, hip circumference, blood pressure and pulse) and data sets as an attached file (please see S1_File.xls).

Comment: I also agree with Reviewer 2 concerning some conclusions that are not supported by the results.

Response: In response to your concerns about the conclusion of the results, we adapted the conclusion section. Specifically, we now provide a more streamlined conclusion based on our results that is consistent with Reviewer 2 recommendations i.e. we deleted the following sentence “Low-threshold interventions have the potential to increase PA and reduce PI in the general elderly population.”

Reviewer #1: I will focus on methods and discussion

Major

1) As the authors acknowledge this is a select population that makes generalisability of the findings questionable.

Response: Thank you for your comment to the generalizability of the findings. Our analysis based on a convenience sample of participants. We used a variety of sampling methods, including the possibility of self-recruitment, some of which have likely increased the proportion of participants with above average PA compared to PA at the population level. We observed high levels of PA, particularly among the older age groups and females, which indicate some selection bias. For this reasons, the composition of our study sample somewhat limits the generalizability of our results.

In our opinion, we already address your concerns about the generalizability of the findings in the limitation section: “Some of these methods have likely increased the proportion of participants with above average PA compared to PA in the general population – so the composition of our study sample somewhat limits the generalizability of our results.” (please see line 386)

2) Some information on the randomisation procedure and how it was implemented is needed.

Response: Thank you for your comment about the random allocation. After screening and accelerometry, participants were individually randomized 1:1 in the intervention and the control group using an automated function (algorithm) in the documentation and study software. The design of the MOVING-study is published (Kleinke et al. 2018). In the trial, we provide information about the randomization process. In response to your concerns about the randomization, we added the following sentence to the manuscript: “The randomization was conducted using an automated algorithm in the documentation software. The documentation software based on electronic case report forms (Kleinke et al. 2018).

3) The sample size has not been arrived to through a power calculation but it is a convenience sample.

Response: The design of the MOVING-study is published (Kleinke et al. 2018). In this paper, we provide detailed information about the power calculation as the following: “The sample size estimate was based on the amount of PA (mean time in min per day), measured in a comparable group of elderly people. In the study in question (Troiano et al.) the amount of PA was on average 15 min of combined PA (moderate and vigorous PA) per day. We took these data as the starting point for the estimation of the required sample size. We assume that participants in the control group remain at approximately the same level while people in the intervention group increase their amount of PA by 20%, which corresponds to an increase from 15 to 18 min of PA per day after 6 months. To demonstrate this effect, a total number of 151 participants is needed (standard deviation 6.0 min, alpha = 0.05, power = 0.80). Assuming a loss to follow-up of about 30%, we will recruit 216 participants.” (Kleinke et al. 2018)

The number of subjects we aimed in the study design (n = 151) was reached (n = 166) in order to prove an effect of the intervention. In the beginning of the methods section, (please see lines 144 - 145), we refer to the design paper of the study for more background information about the methods and study design.

4) Results are poorly reported. Specifically the authors needs to make clear:

a) Drop outs - from 206 to 157?

Response: Thank you for this comment. For the traceability of the dropouts for each study group (with reasons), we provide a Consort diagram (Fig. 1) according to PLOS ONE requirements. The Consort diagram is an attached file (supporting information) – please see Fig. 1 “S1 Fig. Consort diagram: Recruited participants and participants included in the analysis (n = 166)”. Please see also the comment to the adapted sample size in the section for the editor.

b) report characteristics by group (i.e. Table 1). was balance across these key parameters achieved?

Response: Thank you for your constructive comment to the statistical analyses. In response to your concerns about the descriptive statistics, we now provide a comparison between both groups (at baseline) that is consistent with Reviewer 2 recommendations. (please see table 1 “Descriptive characteristics of the study sample (n = 166), complete and valid data sets (baseline, 3 and 6 months follow-up”)

In addition, we checked all variables about significant differences and detected one significant difference in the variable “partnership” (p = 0.006). In the remaining variables, we detected no significant differences between both groups. Due to the significant difference in partnership, we included partnership as a covariable in the Two-Way Mixed ANOVA.

c) if not (i.e. b) then a t-test is not the apprpriate analysis

Response: Thank you for the comment. For baseline data, we analyzed the equality of variables. We considered your concerns about the statistical analyses and adapted our analyses. Specifically, we now performed a Two-Way Mixed ANOVA with repeated measures that is consistent with Reviewer 2 recommendations. In addition, we included partnership as a covariable in the analyses. First, we checked all assumptions and requirements (i.e. assumption of normality) for Mixed ANOVA. Please see a detailed description of the analysis of the assumptions below (Reviewer 2 section). Here, we provide statistical tests and figures for the analysis of the assumptions.

e) a secondary outcome should have been time wearing the device and that should have been reported in the abstract as well

Response: Thank you for this comment. In response to your concerns about the wearing time, we added the following sentence to the abstract: “At baseline, the participants had a mean wearing time of 5,934.5 minutes (SD = 789.5) per week, which corresponds to about 14 hours daily on average.” (please see the result section in the abstract). In addition, we also report wearing time the device (at baseline, 3- and 6-months follow-up) in table 1 (study description). We defined all primary and secondary outcomes of the study a priori. A subsequent adjustment of the secondary outcomes appeared to us to be inappropriate from a methodological point of view.

f) report drop outs by group. Also the number of drop outs is relatively large for a keen sample of healthy adults.

Response: For the traceability of the dropouts for each study group, we provided a Consort diagram (Fig. 1) according to PLOS ONE requirements. The Consort diagram is an attached file (supporting information), in which the number of participants and the Lost to follow up is represent by Baseline, 3 and 6 months follow-up. Please see Fig. 1 “S1 Fig. Consort diagram: Recruited participants and participants included in the analysis (n = 166)”

At baseline, study participants were on average about 71 years old. Although, most participants showed a physically active lifestyle, people (and their partner) of this age group have often underlying diseases. The oldest participant was 85 years old. For this reason, in our opinion the dropout rate in the study is not surprisingly large. We published the study design (Kleinke et al. 2018), in which we assumed a dropout rate of about 30 percent in the power calculation. Due to this, the planned value of the dropouts reflects almost our experiences. In sum, we met the number of participants planned in the power calculation (n = 151) in our published study trial in 2018.

5) No mention of missing data. was multiple imputation considered to boost the sample and get more robust estimates, compared to a complete case analysis?

Response: Thank you for your comment to missing data. After a carefully check of the data, we could increase the number of complete data sets about n = 5. Due to Intention-to-treat principle, we also included n = 4 incomplete datasets. In sum, we could increase the study sample about n = 9, so the study sample in the revised manuscript is now n = 166 at baseline. Due to the small number of incomplete data sets at Follow-up 1 (n = 1) and Follow-up 2 (n = 3), we decided us against an imputation of data. For the main analysis (ANOVA), the number of data sets differ depending on the depended variable, because of the exclusion of outliers.

Reviewer #2: Authors evaluated the effectiveness of a low-threshold intervention to increase PA in elderly. Two groups (intervention-control) were recruited. No differences in most of the variables were found. Authors concluded that PA could be improved in participants with low levels of PA in both groups of this low-threshold intervention. Several concerns regarding the rationale, methodology, results and discussion are explained below.

Authors propose “targeting and tailoring are two important aspects of successful interventions” on basis of the participants’ needs and beliefs, however, what they propose here “a low-threshold intervention” does not meet this criterion. Please justify and show evidence (if the case) of the effectiveness of a more “practicable intervention”, maybe from studies were only biofeedback is used to increase PA.

Response: Thank you for the comment. You are concerns about the design of the intervention are right, our intervention design did not use targeting and tailoring due to the rationale. The rationale of the study was a low-threshold and practicable intervention aiming an increase of PA and a reduction of PI without using complex interventions including telephone/ video calls or personal consultations. In the introduction section, we refer to a systematic review with a positive effect using pedometers. To provide a more streamlined version of the manuscript, we moved this section to the discussion section.

Table 1. Why the number of participants (n) change across the descriptive characteristics, for instance, there are more (161) participants in the line “Number of participants currently living in a partnership (yes)” than those included in the study.

Response: Thank you for your response. The number of participants currently living in a partnership was wrong. Due to the increased number of the study sample (n = 166), we adapted all numbers of the descriptive statistics and removed the mistake. Thank you for that note.

We corrected the mistake and changed the number of participants currently living in a partnership to n = 160. In addition, the difference in the descriptive statistics can be justified by the fact that not every participant answered to all variables in the questionnaire at baseline (i.e. only 160 of 166 participants answered the question about the partnership).

Please, include a Table showing the “descriptive characteristics” between groups and the statistical comparisons to exclude the presence of significant differences in these variables.

Response: Thank you for your constructive comment to the statistical comparison. In response to your concerns about the descriptive statistics, we now provide a comparison between both groups (at baseline) that is consistent with Reviewer 1 recommendations. (please see table 1 “Descriptive characteristics of the study sample (n = 166), complete and valid data sets (baseline, 3 and 6 months follow-up”). In addition, we checked all variables in table 1 on significant differences and detected a significant difference in the variable “partnership” (p = 0.006). Due to the significant difference in partnership, we included partnership as a covariable in the Two-Way Mixed ANOVA to mitigate the effect.

Explain, in the exclusion criteria the cutoff for the fulfillment of the PA according to WHO that you considered. In this regard, if this was an exclusion criterion, why the whole sample was included in the first analysis, and just excluded for the subsequent subgroup analysis?

Response:

Thank you for the comment. For the PA cutoff, we used the age-specific recommendation for people aged ≥65 years from the WHO, which is exactly the age of the study participants (age of ≥65 years). To address your comment, we added the following information to the manuscript about the specific cutoff for the fulfillment of the WHO recommendation: “(≥150 minutes of moderate PA or 75 minutes of vigorous PA).”

We defined all inclusion and exclusion criteria of the study a priori. The self-assessment was about the degree of time of weekly moderate physical activity was included to address a possible ceiling effect of PA in this age group. The operationalization of this exclusion criteria focused on self-reported degree of weekly moderate PA prior to recruitment. In contrast, the analysis focused on the amount of PA objectively measured with the accelerometer device.

Explain, in data analysis, how delta values were calculated.

Response: We followed your recommendation to the statistical analysis and calculated in the revised version of the manuscript a Two-Way ANOVA with repeated measures. Due to the adjustment of the statistical analyses, we have not included delta in the analysis anymore. Due to this, we deleted the analysis of the delta to provide a more streamlined version of the manuscript.

At methodological level, I see two main problems, 1) participants with ≥ 150 minutes of weekly moderate PA at baseline (according to the WHO recommendation) which actually fulfillment an exclusion criterion, were included. 2) The statistical approaching is not justified for this data, why to use t-test instead of two way anova with group (2) and follow-up (3) as factors? The latter could help you to understand within subjects and between groups differences. On the other hand, why data of the first follow-up is not analyzed but included in Tables and Figures?

Response: Thank you for the constructive comments to the statistical analyses that will help to increase the quality of the manuscript.

1) The self-assessment was about the degree of time of weekly moderate physical activity was included to address a possible ceiling effect of PA in this age group. The operationalization of these exclusion criteria focused on self-reported degree of weekly moderate PA prior to recruitment.

We defined all exclusion criteria of the study a priori. We based the main analysis on the whole sample of the study participants irrespective of the results of the measurements within the study. The primary endpoint of our study however was the absolute number of minutes. For the main analysis we did not apply any threshold (i.e. 150 minutes of PA) to avoid any possible bias. However, we nevertheless evaluated the effect of the very active subgroup on the overall intervention effect in a sensitive analysis. In the revised version, we now explicitly report these outcomes. In addition, we discuss the inclusion of the very active subgroup in the limitations. Therefore, we added the following part to the limitations: “Prior to recruitment, we tried to exclude very active people from the study (based on self-assessment) to address a possible ceiling effect. As the results have shown, many participants underestimate their own amount of PA. Therefore, we calculated a sensitive analysis excluding very active participants with an amount of ≥150 minutes of moderate PA. Some ceiling effect reduced the overall intervention effect somewhat.” (please see limitation section)

2) We followed your recommendation to the statistical approach. In the revised version of the manuscript, we calculated a Two-Way Mixed ANOVA with time group (between subject) and time (inner subject) as factors. Due to a significant difference in partnership between intervention and control group at baseline (p = 0.006), we included partnership as a covariable in the Two-Way ANOVA to mitigate a possible effect. In the analysis we detected a significant interaction effect of time*partnership. For the ANOVA, we checked all assumptions and specific requirements. Please see a detail description of the check of the assumptions below. Here, we provide statistical tests for the verification of the assumptions (i.e. Shapiro-Wilk for verification of assumption of normality) and g-g-plots.

Due to adjustment of statistical analysis, we included the 3-months follow-up (first follow-up) in the ANOVA by using time as the independent variable. In addition to the inferential statistics, we analyzed all data using descriptive statistics.

Response: Thank you for the comment to the main analyses. In the revised version of the manuscript, we calculated a mixed ANOVA with time and group as factors and adjusted for partnership. For the ANOVA as a parametric test, crucial assumptions should be checked. Before the main analyses, we first checked the assumptions of the ANOVA beginning with the assumption of normality:

Assumption of normality

First, we checked the assumption of normality for all dependent variables using Shapiro-Wilk Test, because the Shapiro-Wilk-Test has compared to the Kolmogorow-Smirnow-Test a higher statistical power (Razali & Wah, 2011; Steinskog, Tjøstheim & Kvamstø, 2007). In the analyses we detected in two dependent variables an equal distribution of the variables (light and overall PA), and in three dependent variables a non-normal distribution (PI, moderate & vigorous). Light and overall PA were normally distributed for both groups as assessed by the Shapiro-Wilk test (p > .05). Although, assumption of normality is violated in three dependent variables, ANOVA is a robust method against a violation of the assumption (Glass, Peckham, & Sanders, 1972; Harwell, Rubinstein, Hayes, & Olds, 1992; Salkind, 2010). In addition, our sample size is more than n = 30 in each study group. For this reasons, a transformation of the data is not necessary. As a result of the check of the assumptions, we added the following text to the manuscript (please see page 9): “We analyzed the assumptions and requirements of mixed ANOVA i.e. distribution and homoscedasticity and detected no violation of the assumptions”.

Please see attached results of the Shapiro-Wilk-Test for light PA as an example for the analyses of normality. Light PA were normally distributed for both groups as assessed by the Shapiro-Wilk test (p > .05) (please see Tab. 1).

Tab. 1: Shapiro-Wilk-Test for Light PA

In addition to the Shapiro-Wilk-Test, we analyzed the distribution using Q-Q-Plots as you can see in Fig.1 for light PA in the intervention group and Fig. 2 in the control group.

Fig 1: Q-Q-Plot for light PA in the intervention group

Fig 2: Q-Q-Plot for light PA in the control group

In a second example, you can see the result of the Shapiro-Wilk-Test of a non-normal distributed variable. PI was not normally distributed for both groups as assessed by the Shapiro-Wilk test (p < .05) (please see Tab. 2).

Tab. 2: Shapiro-Wilk-Test for PI

Literature

1. Glass, G. V., Peckham, P. D., & Sanders, J. R. (1972). Consequences of Failure to Meet Assumptions Underlying the Fixed Effects Analyses of Variance and Covariance. Review of Educational Research, 42(3), 237–288. doi:10.3102/00346543042003237

2. Harwell, M. R., Rubinstein, E. N., Hayes, W. S., & Olds, C. C. (1992). Summarizing Monte Carlo Results in Methodological Research: The One- and Two-Factor Fixed Effects ANOVA Cases. Journal of Educational and Behavioral Statistics, 17(4), 315–339. doi:10.3102/10769986017004315

3. Salkind, N. J. (2010). Encyclopedia of Research Design (Vol. 2). Los Angeles: Sage.

4. Razali, N. M., & Wah, Y. B. (2011). Power comparisons of Shapiro-Wilk, Kolmogorov-Smirnov, Lilliefors and Anderson-Darling tests. Journal of Statistical Modeling and Analytics, 2(1), 21-33.

5. Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52(3/4), 591-611.

6. Steinskog, D. J., Tjøstheim, D. B., & Kvamstø, N. G. (2007). A cautionary note on the use of the Kolmogorov-Smirnov test for normality. Monthly Weather Review, 135(3), 1151-1157.

Check for outliers in the data set

In a second step, we systematically detected outliers (for each dependent variable). Outliers were defined as any data point that is more than 1.5 times (light outliers) or more than 3 times of the interquartile range (IQR) (extreme outliers). We excluded outliers (data points > than 1.5 times of the IQR) from the analyses, because ANOVA as a parametric test is not robust against outliers. In addition, we wanted to evaluate the intervention effect in the general population. In our study sample, the amount of PA is already overestimated (i.e. one participant achieved about 800 minutes moderate PA per week). The number of excluded datasets is little.

Due to the check of assumptions, we excluded different datasets regarding the dependent variable. Thus, the number of included data sets slightly changed between the Two-Way Mixed ANOVAS.

After excluding outliers, the depending variables PI, light and overall PA were normally distributed. In addition, we also checked the homogeneity of the error variances (assessed by Levene’s test) and covariances (by Box’s test) and detected no violation of the assumptions.

The only significant result was a difference in delta value between groups, however, this result did not confirm a PA increase due the intervention, but a decrease in the PA in the control group, according with Figure 2, is that right? Therefore, the effectiveness of the intervention is not corroborated?

Response: In the revised version of the manuscript, we changed the statistical method according to your recommendations. Therefore, we have not evaluated delta values any more and deleted it from the manuscript.

In the first paragraph of the discussion, authors stated that the intervention group showed more overall PA compared to the participants in the control group, however this result was not significant. Therefore, no differences between groups were observed at any time.

Response: Thank you for the comment. Please see comment below.

In lines 286-287 authors suggest “Interventions should focus on the target group and especially on individual factors (e.g. age, sex)”, but not analysis was carried out with these variables in this study, it could be an interesting approaching. It is known, in previous literature, the differences between sexes in terms of PA and self-efficacy to PA.

Response: In a previous paper, we already analysed the determinants for PA and PI in a group of healthy elderly people separated by sex (Kleinke, 2020). Here, we found i.e. a statistical significant positive effect of self-efficacy in women (p = 0.020). Due to significant group difference, we included partnership as a covariate in the ANOVA. To address your concerns about the intervention design, we added the following text to the manuscript: “As previous research has shown”. (please see line 350 - 351) In this paper, we wanted to evaluate a low-threshold intervention to improve overall PA and reduce PI.

From lines 301-313 authors discuss the importance of light PA increase and PI decrease, but their results did not show evidence of the effectiveness of the intervention even for these levels of PA.

Response: Although, there was no significant intervention effect on light PA, people aged over 65 years are a relevant target group especially for light PA. Independently and as the descriptive results show, study participants spent most of the time in light PA. Thus, interventions for the elderly should especially focus on light PA.

In the conclusion authors suggest “Low-threshold interventions have the potential to increase PA and reduce PI in the general elderly population” however their results did not support this statement.

Response: Thank you for the comment. We deleted the sentence from the manuscript and tried to focus our statements which are supported by the results. In the revised version, we addressed your comment as the following: “Our analyses shows limited and not significant, but seems to have a small positive effect of a low-threshold intervention after 6 months follow-up.” (please see conclusion section).

Literature (1–4)

1. Stamatakis E, Ekelund U, Ding D, Hamer M, Bauman AE, Lee I-M. Is the time right for quantitative public health guidelines on sitting? A narrative review of sedentary behaviour research paradigms and findings. Br J Sports Med [Internet]. 2018; Available from: https://bjsm.bmj.com/content/early/2018/06/08/bjsports-2018-099131

2. Kleinke F, Schwaneberg T, Weymar F, Penndorf P, Ulbricht S, Lehnert K, et al. MOVING: Motivation-Oriented interVention study for the elderly IN Greifswald: study protocol for a randomized controlled trial. Trials. 2018 Jan;19(1):57.

3. Kleinke F, Penndorf P, Ulbricht S, Dörr M, Hoffmann W, van den Berg N. Levels of and determinants for physical activity and physical inactivity in a group of healthy elderly people in Germany: Baseline results of the MOVING-study. PLoS One [Internet]. 2020 Aug 13;15(8):e0237495. Available from: https://doi.org/10.1371/journal.pone.0237495

4. Ekelund U, Tarp J, Steene-Johannessen J, Hansen BH, Jefferis B, Fagerland MW, et al. Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ [Internet]. 2019;366. Available from: https://www.bmj.com/content/366/bmj.l4570

27 May 2021

PONE-D-21-04633R1

A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study

PLOS ONE

Dear Dr. Kleinke,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Very few corrections are missing for the acceptance of the article. Please provide the meaning of each line color in Figure 2, correct the use of parentheses and square brackets in the results section, and complete Reviewer 2's suggestions where possible.

Please submit your revised manuscript by June 7th 2021. I await your response very soon! Although if you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Thalia Fernandez, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments (if provided):

Dear Dr. Kleinke,

Very few corrections are missing for the acceptance of the article. Provide the meaning of each line color in Figure 2, correct the use of parentheses and square brackets in the results section, and complete Reviewer 2's suggestions where possible.

I await your response very soon!

Kind regards,

Thalía

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I am happy with the authors' responses. Some minor comments and perhaps advice for future work:

1) try to use a regression model over ANOVA. it is more flexible, easier to interpret and allows for controlling for numerous covariates.

baseline characteristics appear similar across the two groups. report as mean (95% CI) rather than M (95% CI) in the respective table. You could include comparisons at baseline between the two groups, although I don't think anything is significantly different.

sensitivity rather than sensitive analyses

in the absence of a main effect we don't tend to examine for interactions

Reviewer #3: Minor details:

The authors should provide the meaning of each line color in figure 2, and correct the usage of parentheses and brackets in the results section.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #3: Yes: Graciela Catalina Alatorre Cruz

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

31 May 2021

PONE-D-21-04633

A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study

PLOS ONE

Response to Reviewers

Thank you for coordinating the review process and for your constructive comments to the manuscript. In our opinion, the comments help to increase the quality of the manuscript, especially the part of the statistical analyses. We have revised the manuscript again and addressed all reviewer’s comments from the minor revision.

Reviewer #1

Comment: I am happy with the authors' responses. Some minor comments and perhaps advice for future work:

1) try to use a regression model over ANOVA. it is more flexible, easier to interpret and allows for controlling for numerous covariates.

Response: Thank you for this comment to the statistical analysis. In the revised version of the manuscript, we provide a Two-Way Mixed ANOVA with repeated measures that is in line with reviewer 2 from the major revision. For further analysis, we will take into account the advice from reviewer 1 and consider using a regression model.

Comment: baseline characteristics appear similar across the two groups. report as mean (95% CI) rather than M (95% CI) in the respective table. You could include comparisons at baseline between the two groups, although I don't think anything is significantly different.

Response: At baseline, we checked all variables regarding significant group differences. Here, we detected one systematic difference in the variable “partnership” between intervention and control group. Therefore, we added the variable “partnership” as a covariate in the Two-Way Mixed ANOVA to mitigate the effect. All other variables in table 1 were not significantly different. In the revised version of the manuscript, we added the following sentence: “For baseline data, we analyzed the equality of variables between both study groups.” (please see lines: 210 – 211).

We changed the report of table 2 that is consistent with Reviewer 1 recommendations (Mean (95% CI - please see green color in table 2).

Comment: sensitivity rather than sensitive analyses

Response: Thank you for this comment. In the revised version of the manuscript, we addressed the comment.

Comment: in the absence of a main effect we don't tend to examine for interactions

Response: Thank you for this comment. We followed your recommendation and adapted the section with very few changes without reporting the details about the interaction effect. Now, we summarize the details about the interaction effect in one sentence: “P-values were ranging between 0.317 to 0.546.”. (Please see line: 260).

Reviewer #3

Comment: The authors should provide the meaning of each line color in figure 2, and correct the usage of parentheses and brackets in the results section.

Response: Thank you for this comment. The blue line represents the intervention group and the red line represents participants in the control group. We also added the meaning of each line in the manuscript (line 247). In the revised version of the manuscript, we have addressed Reviewer 3 comments and corrected the usage of parentheses and brackets in the results section (please see changes in green color and review mode in the manuscript).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

24 Jun 2021

PONE-D-21-04633R2

A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study

PLOS ONE

Dear Dr. Kleinke,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

I consider that the article could be accepted after it satisfies the following requests:

1.- The authors should provide the meaning of each line color in figure 2.

2.- The authors should correct the usage of parentheses and brackets in the results section.

3.- Please report as mean (95% CI) rather than M (95% CI) in the respective table.

4.- Please include comparisons at baseline between the two groups.

5.- Please, modify your wording: "sensitivity" rather than "sensitive analyses"

Please submit your revised manuscript by July 3, 2021. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Thalia Fernandez, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

6 Aug 2021

PONE-D-21-04633

A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study

PLOS ONE

Response to Reviewers

Thank you for coordinating the review process and for your constructive comments to the manuscript. In our opinion, the comments help to increase the quality of the manuscript, especially the part of the statistical analyses. We have revised the manuscript again and addressed all reviewer’s comments from the minor revision.

Reviewer #1

Comment: I consider that the article could be accepted after it satisfies the following requests:

1. The authors should provide the meaning of each line color in figure 2.

Response: The blue line represents the intervention group and the red line represents participants in the control group. We also added the meaning of each line in the manuscript (please see lines 247, 249 and 608/609 in “manuscript_with_track_changes”).

Comment: 2. The authors should correct the usage of parentheses and brackets in the results section.

Response: In the revised version of the manuscript, we have addressed Reviewer 3 comments and corrected the usage of parentheses and brackets in the results section (please see changes in green color and review mode in the manuscript).

Comment: 3. Please report as mean (95% CI) rather than M (95% CI) in the respective table.

Response: In the revised version of the manuscript, we changed the report of Mean (95% CI) that is consistent with Reviewer 1 recommendations. Please see Table 2 in the manuscript.

Comment: 4. Please include comparisons at baseline between the two groups.

Response: At baseline, we checked all variables regarding significant group differences. Here, we detected one systematic difference in the variable “partnership” between intervention and control group. Therefore, we added the variable “partnership” as a covariate in the Two-Way Mixed ANOVA to mitigate the effect. Other variables in table 1 were not significantly different. In the revised version of the manuscript, we added the following sentence: “For baseline data, we analyzed the equality of variables between both study groups.” (please see lines: 210 – 211).

Comment: 5. Please, modify your wording: "sensitivity" rather than "sensitive analyses"

Response: Thank you for this comment. In the revised version of the manuscript, we addressed the comment.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

31 Aug 2021

A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study

PONE-D-21-04633R3

Dear Dr. Kleinke,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Thalia Fernandez, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

3 Sep 2021

PONE-D-21-04633R3

A low-threshold intervention to increase physical activity and reduce physical inactivity in a group of healthy elderly people in Germany: Results of the randomized controlled MOVING study

Dear Dr. Kleinke:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Thalia Fernandez

Academic Editor

PLOS ONE