Wear Time and Valid Day Calculator.
Set valid wear rules for accelerometer and wearable studies, and see how many days of monitoring give you usable data.
Plan the monitoring period from the share of days you expect to be valid, allowing for differences between participants, or check one participant's daily wear against a published rule. Everything is calculated in your browser.
How to Use the Wear Time Calculator
- Choose a valid wear rule. Start from a published rule or set your own: minimum hours of wear for a valid day, minimum valid days, and any weekend days required.
- To plan, enter the monitoring length, the first day and the share of days you expect to be valid, ideally from pilot data. In 87 adults who wore a hip accelerometer for 21 days, 78%, 67% and 52% of days reached 8, 10 and 12 hours of wear[1].
- Keep differences between participants at 0.25 unless pilot data say otherwise, so the result allows for some participants missing many days.
- Set the share of participants who must have usable data. The planner finds the shortest monitoring length that reaches it, and how many to enrol.
- To check a participant, enter or paste the hours of wear for each day. You see which days are valid, whether the participant passes and why, and can download the result as CSV.
Valid day rules are one part of device data quality, which our guide to wearable data quality in clinical trials covers more broadly. For running the study itself, see how WeGuide supports wearable device studies.
Valid Day Rules and Where They Come From
A rule has two parts: a minimum amount of wear for a day to count, and a minimum number of valid days for a participant to count.
| Rule | Valid day | Valid participant | Setting and source |
|---|---|---|---|
| NHANES 2003 to 2006 | At least 10 hours | At least 4 valid days | Hip, waking hours, US participants aged 6 and over[2] |
| ISCOLE | At least 10 hours of waking wear | At least 4 valid days, including 1 weekend day | Waist, 24 hour protocol, children aged 9 to 11 in 12 countries[3] |
| Migueles 2017 review | At least 10 hours per awake period | At least 4 valid days | Recommendation from a review of 235 ActiGraph GT3X studies[4] |
| Women's Health Study | At least 10 hours | At least 4 days | Hip, waking hours, older US women[5] |
| Large tracker studies | At least 10 hours | At least 3 valid days a week | Most common choices in 27 studies with 1,000 or more participants[6] |
| Millennium Cohort Study | At least 10 hours | At least 2 valid days | Hip, waking hours, UK children aged 7[7] |
| ALSPAC | At least 10 hours | At least 3 valid days | UK children aged 11[7][8] |
| COPD rehabilitation | At least 8 hours | At least 4 weekdays, weekends not counted | SenseWear armband, waking hours[9] |
| GGIR, 24 hour wrist | At least 16 valid hours in a calendar day | Not set by GGIR | Default day criterion for raw wrist data[10] |
| UK Biobank | No daily rule | At least 72 hours of wear, and wear in every hour of the 24 hour cycle | Dominant wrist, 24 hour protocol, device sent by post[11] |
| All of Us | At least 10 hours of wear and 100 steps | Months with fewer than 15 valid days excluded | Participants' own Fitbit devices[12] |
| Sleep actigraphy | A usable night, as defined by the study | At least 5 usable nights, recording at least a full week | Children and adolescents[13] |
All rules except sleep are presets in the calculator. UK Biobank is in the checker only, which checks its 72 hour total, because hourly coverage needs minute data. We found no published valid day rule developed for smart rings, so none is offered.
How the Planner Works
If every participant has the same chance p of a valid day and days are independent, the number of valid days in an n day window is binomial, and the share of participants with at least k valid days is:
P = Σ from j = k to n of C(n, j) × pj × (1 − p)n − j
Weekend rules need the calendar. A 7 day window always holds 2 weekend days, but a 10 day window holds 2, 3 or 4, depending on the first day. The planner treats valid weekdays X and valid weekend days Y as separate counts, each with its own probability. A participant meets the rule when X + Y is at least k and Y is at least w:
P = Σ over y ≥ w of P(Y = y) × P(X ≥ k − y)
An optional daily decline lowers the probability by a fixed number of percentage points each day. The shortest monitoring length is the first window, counting up from 1 day, that reaches your target. Adding a day can only add valid days, so the share never falls as the window grows.
Why Differences Between Participants Matter
The binomial model assumes a missed day is equally likely for everyone. In real cohorts missed days cluster in some participants, which makes lenient rules harder to meet than the binomial model suggests.
Among 1,296 Danish adolescents, 93.1% kept at least 4 valid days of 10 hours, where the best fitting binomial model predicts 99.7%[14]. A beta binomial model, in which each participant has their own chance of a valid day, fits all seven published shares, from 1 to 7 valid days, with a between participant correlation of 0.26 (WeGuide calculation).
In 87 adults monitored for 21 days, 67% of days reached 10 hours. Independence predicts that 83% of weeks would hold at least 4 valid days, but 74% did[1]. The correlation that reproduces 74% is 0.21 (WeGuide calculation).
The planner therefore uses the beta binomial model with a default correlation ρ of 0.25 and shows the binomial result beside it. With 80% of days valid and the NHANES rule over 7 days, they give 86.9% and 96.7%.
q ~ Beta(α, β), with mean p and ρ = 1 ÷ (α + β + 1)
P = Σ C(D, x) × C(E, y) × B(α + x + y, β + n − x − y) ÷ B(α, β)
The sum runs over the valid weekday counts x and weekend counts y that meet the rule, with D weekdays and E weekend days in the window. This model needs one probability for all days and no decline, so with a separate weekend share or a decline the planner falls back to the binomial model.
Choosing a Rule That Fits Your Study
Stricter rules give a more reliable estimate per participant but keep fewer participants. Migueles and colleagues recommend testing several criteria for the best balance of sample size and reliability[4].
The daily threshold
Shorter days add error. In a semisimulation of 1,000 NHANES days, absolute percent error against a 14 hour day was 3.9 to 14.1% at 13 hours of wear and 24.6 to 40.3% at 10 hours[15]. Non-wear time that is not removed tends to be counted as sedentary time[4].
The number of valid days
More days shrink the sample. Among Danish adolescents, 93.1% had at least 4 valid days of 10 hours, but only 57.2% had 7[14].
Weekend days
A weekend requirement depends on the outcome. Requiring weekend data did not improve reliability in children aged 7[7], and leaving weekends out reduced the sample size needed for steps in COPD rehabilitation from 83 to 56[9].
Who gets excluded
Missing wear is rarely random. Older adolescents and those with a higher BMI were more likely to be excluded under a 10 minute non-wear rule[14]. Among 3,601 adults in Hong Kong, older people and those with a full time job, tertiary education, no smoking and better self rated health wore devices longer[16].
How Many Days Give a Reliable Estimate
Valid days for a reliability of about 0.80, unless stated. Review consensus is 3 to 5 days in adults and 4 to 9 in children and adolescents[17]. Because not every day is valid, monitoring needs to run longer.
| Population and measure | Valid days needed | Source |
|---|---|---|
| Children in grades 1 to 6, MVPA | 4 to 5 | Trost 2000[18] |
| Adolescents in grades 7 to 12, MVPA | 8 to 9 | Trost 2000[18] |
| Adults, activity counts and MVPA | 3 to 4, and at least 7 for inactivity | Matthews 2002[19] |
| Adults over 21 days, hip | 7 to 10 for overall activity and MVPA, 3 to 4 for light activity, 13 to 16 for sedentary minutes | Aadland 2015[1] |
| Older adults | 3 for activity, 5 for sedentary time | Hart 2011[20] |
| Middle aged adults, wrist | 6, including Saturday and Sunday, for all intensities | Dillon 2016[21] |
| Adults, pedometer steps | Any 3 days | Tudor-Locke 2005[22] |
| Sleep, children and adolescents | 5 or more usable nights | Acebo 1999[13] |
| Sleep, working adults | At least 5 nights for sleep efficiency, more than 7 for total sleep time | Aili 2017[23] |
Participants found 21 days lengthy. The authors suggested 14 days may be enough if participants give 4 to 5 valid days a week[1].
Consumer and Research Devices Define Wear Differently
Research accelerometers record counts or raw acceleration, so wear is derived with a documented non-wear rule. Count based rules include Troiano and Choi[2][24], raw signal rules include the GGIR and UK Biobank methods[10][11], and the SenseWear armband detects skin contact[9]. Our research reference for ActiGraph devices covers the LEAP and wGT3X-BT in more depth.
In the consumer device studies reviewed for this tool, the device did not supply a non-wear algorithm, so researchers inferred wear from the data: hours with any heart rate data[25], at least one heart rate value in each 15 minute window[26], or hours with step counts above zero[12]. Supported Garmin models are an exception, with raw accelerometer data through the Garmin Health SDK, which the WeGuide Garmin integration collects at 25 Hz.
The definition alone changes the answer. On the same Fitbit Charge 2 data from 186 patients over 90 days, median usage was 90.0% by heart rate hours, 83.7% by heart rate minutes, 77.1% with a fixed 8 hour sleep assumption, and 43.8% by the count based NHANES rule with no sleep assumption[25].
Steps alone cannot tell sitting still from not wearing the device. See how watches track steps and what daily step count shows as a digital biomarker. Fix the wear definition first, then the threshold, and report both.
What Improves Wear Time
- Placement and handout. Across 123 data collection waves in adults, wrist wear raised the share meeting minimum wear criteria by 14 percentage points against waist wear. Handing devices out in person raised adherence to minimum wear by 15 points against sending them by post[27].
- 24 hour protocols. In US children, a 24 hour waist protocol gave 61.8 more minutes a day of waking wear than the NHANES waking protocol, and 76.7% against 62.6% valid cases under the same rule[3]. Children also wore wrist devices more than hip devices under every wear criterion tested[28].
- Payment for complete days. In four high schools, each using one strategy, 96% of students met at least 4 of 7 complete days when paid for each complete day, against 70% in the control school[29]. Allocation by school limits what this shows about cause.
- Real time monitoring with site follow up. In the LIBERTAS phase 3 trial, mean wear rose from under 19 to nearly 22 hours a day after a sponsor run workflow began flagging participants below 75% adherence to their sites[30]. This was a sponsor authored before and after comparison, not a randomised test.
- Reminders. Among 375 young people, wear time did not differ between simple and rationale giving SMS reminders, nor between those who opted in to reminders and those who did not, a comparison that was not randomised[31].
Limitations to Keep in Mind
- The correlation is an estimate from two published cohorts. Use pilot data where you can.
- No validated daily decline exists. Wear decreased over time among 623 people with recurrent depression followed for a median of 541 days[26], but there is no verified rate for a 7 to 14 day protocol. Treat the decline field as a scenario.
- Completeness is not lack of bias. A high usable share says nothing about who was lost, and exclusion is linked to age, BMI and health[14][16].
- Posted devices need a protocol window. Algorithms alone counted days in transit as wear days[32]. Enter protocol days only.
Running a device study?
WeGuide collects wearable data from Garmin devices, including raw accelerometer and gyroscope data at 25 Hz from supported models through the Garmin Health SDK, and from Apple Health and Android Health Connect. Participant tracking shows each person's adherence as data arrives, so your team can follow up early.
Frequently Asked Questions
Answers about valid wear rules, monitoring length and wear time in accelerometer and wearable studies.
Have another question?How many hours of wear make a valid day?
Most studies count a day as valid with at least 10 hours of waking wear, the most common rule in large tracker studies and the rule in NHANES 2003 to 2006. For 24 hour wrist data processed in GGIR, the default is 16 valid hours. COPD trials have used 8 hours. Shorter thresholds keep more days but add error, so choose the highest threshold your participants can reach.
How many valid days do I need?
Most accelerometer studies require at least 3 or 4 valid days. Three days is the most common rule in large tracker studies, and a review of ActiGraph studies recommends at least 4. Reliable estimates of habitual activity need about 3 to 5 days in adults and 4 to 9 days in children and adolescents. The monitoring period has to be longer, because not every day is valid.
Do I need a weekend day?
Only if your outcome or population calls for it. Children were more active at weekends and adolescents less active, so some studies require a weekend day. Yet in children aged 7, requiring weekend data did not improve reliability, and in COPD rehabilitation leaving weekends out reduced the sample size needed. Treat it as a design choice to justify in the protocol.
Should participants wear the device for 24 hours or only while awake?
A 24 hour protocol usually gives more wear time. In US children, a 24 hour waist protocol gave 61.8 more minutes of waking wear per day than the NHANES waking protocol, and more valid cases under the same rule. With a 24 hour protocol, set the daily threshold for the longer day, such as the GGIR default of 16 valid hours.
How long should the monitoring period be?
Long enough for your valid day rule to reach the share of participants you need. Seven days is the most common period, but it is not always enough. With the ISCOLE rule and 70% of days valid, 10 days are needed for 90% of participants to qualify, by WeGuide calculation. Sleep actigraphy in young people needs at least a full week.
What share of participants will have usable data?
It depends on the rule, the protocol and the population, and published shares range from about 57% to 96%. Among Danish adolescents, 93.1% had at least 4 valid days but only 57.2% had 7. Under a 24 hour waist protocol, 76.7% of US children met the ISCOLE rule. Use the planner with your own share of valid days to estimate it.
Why does the calculator include differences between participants?
Because missed days cluster in some participants, and a model that ignores this overstates how many people meet a lenient rule. In Danish adolescents, the binomial model predicted 99.7% with at least 4 valid days, against 93.1% observed. A model with a between participant correlation of about 0.26 fitted closely. The default of 0.25 reflects fits to two published cohorts.
How do smartwatches and rings define wear time?
Studies infer wear from the data a consumer device records, such as hours with heart rate data or hours with steps. The choice matters: on the same Fitbit data, median usage was 90.0% by heart rate hours and 43.8% by a count based rule. We found no published valid day rule developed for smart rings, so report your definition and threshold in full.
Which non-wear algorithm should I use?
For count data, use a published rule such as Choi, which was validated in a room calorimeter and agreed best with participant diaries in a large cohort of older women, or Troiano for comparison with NHANES. For raw data, use the rule of the processing pipeline you cite, such as the UK Biobank or GGIR method. Report the algorithm with the hour threshold.
Do reminders improve wear time?
The evidence for reminders is weak. Among 375 young people, wear time did not differ between simple and rationale giving SMS reminders, nor between those who chose reminders and those who did not. Better evidence supports protocol choices such as wrist wear, handing devices out in person and 24 hour wear, and paying participants for each complete day.
Related Research Tools
Free tools for planning and running clinical research, built on published methods.
References
- Aadland E, Ylvisåker E. Reliability of Objectively Measured Sedentary Time and Physical Activity in Adults. PLoS One. 2015;10(7):e0133296. PubMed 26192184
- Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008;40(1):181-8. PubMed 18091006
- Tudor-Locke C, Barreira TV, Schuna JM Jr, et al. Improving wear time compliance with a 24-hour waist-worn accelerometer protocol in the International Study of Childhood Obesity, Lifestyle and the Environment (ISCOLE). Int J Behav Nutr Phys Act. 2015;12:11. PubMed 25881074
- Migueles JH, Cadenas-Sanchez C, Ekelund U, et al. Accelerometer Data Collection and Processing Criteria to Assess Physical Activity and Other Outcomes: A Systematic Review and Practical Considerations. Sports Med. 2017;47(9):1821-1845. PubMed 28303543
- Lee IM, Shiroma EJ, Kamada M, et al. Association of Step Volume and Intensity With All-Cause Mortality in Older Women. JAMA Intern Med. 2019;179(8):1105-1112. PubMed 31141585
- Chan A, Chan D, Lee H, et al. Reporting adherence, validity and physical activity measures of wearable activity trackers in medical research: A systematic review. Int J Med Inform. 2022;160:104696. PubMed 35121356
- Rich C, Geraci M, Griffiths L, et al. Quality control methods in accelerometer data processing: defining minimum wear time. PLoS One. 2013;8(6):e67206. PubMed 23826236
- Mattocks C, Ness A, Leary S, et al. Use of accelerometers in a large field-based study of children: protocols, design issues, and effects on precision. J Phys Act Health. 2008;5 Suppl 1:S98-111. PubMed 18364528
- Demeyer H, Burtin C, Van Remoortel H, et al. Standardizing the analysis of physical activity in patients with COPD following a pulmonary rehabilitation program. Chest. 2014;146(2):318-327. PubMed 24603844
- GGIR documentation. GGIR parameters, version 3.3-9. Accessed 20 October 2026. wadpac.github.io/GGIR
- Doherty A, Jackson D, Hammerla N, et al. Large Scale Population Assessment of Physical Activity Using Wrist Worn Accelerometers: The UK Biobank Study. PLoS One. 2017;12(2):e0169649. PubMed 28146576
- Master H, Annis J, Huang S, et al. Association of step counts over time with the risk of chronic disease in the All of Us Research Program. Nat Med. 2022;28(11):2301-2308. PubMed 36216933
- Acebo C, Sadeh A, Seifer R, et al. Estimating sleep patterns with activity monitoring in children and adolescents: how many nights are necessary for reliable measures? Sleep. 1999;22(1):95-103. PubMed 9989370
- Toftager M, Kristensen PL, Oliver M, et al. Accelerometer data reduction in adolescents: effects on sample retention and bias. Int J Behav Nutr Phys Act. 2013;10:140. PubMed 24359480
- Herrmann SD, Barreira TV, Kang M, et al. Impact of accelerometer wear time on physical activity data: a NHANES semisimulation data approach. Br J Sports Med. 2014;48(3):278-82. PubMed 22936409
- Lee PH, Macfarlane DJ, Lam TH. Factors associated with participant compliance in studies using accelerometers. Gait Posture. 2013;38(4):912-7. PubMed 23688408
- Trost SG, McIver KL, Pate RR. Conducting accelerometer-based activity assessments in field-based research. Med Sci Sports Exerc. 2005;37(11 Suppl):S531-43. PubMed 16294116
- Trost SG, Pate RR, Freedson PS, et al. Using objective physical activity measures with youth: how many days of monitoring are needed? Med Sci Sports Exerc. 2000;32(2):426-31. PubMed 10694127
- Matthews CE, Ainsworth BE, Thompson RW, et al. Sources of variance in daily physical activity levels as measured by an accelerometer. Med Sci Sports Exerc. 2002;34(8):1376-81. PubMed 12165695
- Hart TL, Swartz AM, Cashin SE, et al. How many days of monitoring predict physical activity and sedentary behaviour in older adults? Int J Behav Nutr Phys Act. 2011;8:62. PubMed 21679426
- Dillon CB, Fitzgerald AP, Kearney PM, et al. Number of Days Required to Estimate Habitual Activity Using Wrist-Worn GENEActiv Accelerometer: A Cross-Sectional Study. PLoS One. 2016;11(5):e0109913. PubMed 27149674
- Tudor-Locke C, Burkett L, Reis JP, et al. How many days of pedometer monitoring predict weekly physical activity in adults? Prev Med. 2005;40(3):293-8. PubMed 15533542
- Aili K, Åström-Paulsson S, Stoetzer U, et al. Reliability of Actigraphy and Subjective Sleep Measurements in Adults: The Design of Sleep Assessments. J Clin Sleep Med. 2017;13(1):39-47. PubMed 27707448
- Choi L, Liu Z, Matthews CE, et al. Validation of accelerometer wear and nonwear time classification algorithm. Med Sci Sports Exerc. 2011;43(2):357-64. PubMed 20581716
- Speier W, Dzubur E, Zide M, et al. Evaluating utility and compliance in a patient-based eHealth study using continuous-time heart rate and activity trackers. J Am Med Inform Assoc. 2018;25(10):1386-1391. PubMed 29850807
- Matcham F, Leightley D, Siddi S, et al. Remote Assessment of Disease and Relapse in Major Depressive Disorder (RADAR-MDD): recruitment, retention, and data availability in a longitudinal remote measurement study. BMC Psychiatry. 2022;22(1):136. PubMed 35189842
- Pulsford RM, Brocklebank L, Fenton SAM, et al. The impact of selected methodological factors on data collection outcomes in observational studies of device-measured physical behaviour in adults: A systematic review. Int J Behav Nutr Phys Act. 2023;20(1):26. PubMed 36890553
- Fairclough SJ, Noonan R, Rowlands AV, et al. Wear Compliance and Activity in Children Wearing Wrist- and Hip-Mounted Accelerometers. Med Sci Sports Exerc. 2016;48(2):245-53. PubMed 26375253
- Sirard JR, Slater ME. Compliance with wearing physical activity accelerometers in high school students. J Phys Act Health. 2009;6 Suppl 1(Suppl 1):S148-55. PubMed 19998861
- Del Campo AO, Sanchez D, et al. Improving actigraphy adherence among clinical trial participants with a real-time data-driven workflow. BMC Digital Health. 2025;3:88. doi:10.1186/s44247-025-00213-0
- Heino MTJ, Knittle K, Haukkala A, et al. Simple and rationale-providing SMS reminders to promote accelerometer use: a within-trial randomised trial comparing persuasive messages. BMC Public Health. 2018;18(1):1352. PubMed 30526616
- Keadle SK, Shiroma EJ, Freedson PS, et al. Impact of accelerometer data processing decisions on the sample size, wear time and physical activity level of a large cohort study. BMC Public Health. 2014;14:1210. PubMed 25421941
Formulas checked against an independent reference implementation and the original publications on 20 October 2026 by the WeGuide editorial team. This tool is for research planning and education. It is not a medical device and does not give medical advice.
