Urolithiasis (2018) 46:343–348
https://doi.org/10.1007/s00240-017-1006-x
ORIGINAL PAPER
Accuracy of daily fluid intake measurements using a “smart”
water bottle
Michael S. Borofsky1 · Casey A. Dauw2 · Nadya York3 · Colin Terry4 ·
James E. Lingeman3
Received: 19 April 2017 / Accepted: 15 September 2017 / Published online: 4 October 2017
© Springer-Verlag GmbH Germany 2017
Abstract High fluid intake is an effective preventative difference between hand and smart bottle measurements was
strategy against recurrent kidney stones but is known to 0.0% (95% CI − 3%, 3%). Intraclass correlation coefficient
be challenging to achieve. Recently, a smart water bottle (ICC), calculated to assess consistency between hand meas-
(Hidrate Spark™, Minneapolis, MN) was developed as a ures and bottle measures, was 0.97 (0.95, 0.98) indicating
non-invasive fluid intake monitoring system. This device an extremely high consistency between measures. 24-h fluid
could help patients who form stones from low urine volume intake measurements from a novel fluid monitoring system
achieve sustainable improvements in hydration, but has yet (Hidrate Spark™) are accurate to within 3%. Such technol-
to be validated in a clinical setting. Hidrate Spark™ uses ogy may be useful as a behavioral aide and/or research tool
capacitive touch sensing via an internal sensor. It calculates particularly among recurrent stone formers with low urinary
volume measurements by detecting changes in water level volume.
and sends data wirelessly to users’ smartphones through an
application. A pilot study was conducted to assess accu- Keywords Metabolic stone · Urolithiasis · Mobile
racy of measured fluid intake over 24 h periods when used health · Technology · Smartphone · Fluid · Water ·
in a real life setting. Subjects were provided smart bot- Nephrolithiasis
tles and given short tutorials on their use. Accuracy was
determined by comparing 24-h fluid intake measurements
calculated through the smart bottle via sensor to standard Introduction
volume measurements calculated by the patient from hand
over the same 24 h period. Eight subjects performed sixty- Urinary stone disease is one of the most common and costly
two 24-h measurements (range 4–14). Mean hand measure- diseases in medicine affecting nearly 10% of the US popu-
ment was 57.2 oz/1692 mL (21–96 oz/621–2839 mL). Cor- lation [1]. Prevalence is growing rapidly with rates having
responding mean smart bottle measurement underestimated more than doubled over only the past 15 years, making it
true fluid intake by 0.5 ozs. (95% CI −1.9, 0.9). Percent the most costly urologic condition [1, 2]. One of the biggest
obstacles in controlling the growing incidence of stone dis-
ease is better preventative care, especially considering risk
* James E. Lingeman of recurrence is as high as 50% within 5 years [2]. To date,
jlingeman@iuhealth.org
there are few treatments with strong evidence supporting
1
Department of Urology, University of Minnesota, their use in stone prevention. High fluid intake is perhaps
Minneapolis, MN, USA the most effective preventative strategy based on a 5-year
2
Department of Urology, University of Michigan, Ann Arbor, randomized control study which demonstrated a 12% recur-
MI, USA rence rate among patients encouraged to achieve a goal of
3
Department of Urology, Indiana University School 2 L of urine per day compared to a 27% recurrence in the
of Medicine, 1801 North Senate Blvd., Suite 220, group not encouraged to increase fluid intake [3]. High fluid
Indianapolis, IN 46202, USA intake is also a recommended prevention strategy by both
4
Methodist Research Institute, Indianapolis, IN, USA the American Urological Association (AUA) and American
13
Vol.:(0123456789)
, 344 Urolithiasis (2018) 46:343–348
College of Physicians (ACP) who advocate daily goal urine Methods
volume of 2.5 L for all stone formers [3, 4].
Achieving significant increase in hydration and subse- Hidrate Spark™ (Fig. 1) uses capacitive touch sensing via a
quently urinary volume is a considerable clinical challenge sensor extending from the lid to the base of the bottle which
for patients. While there is little data assessing patient com- calculates fluid volume measurements by detecting changes
pliance specifically for fluid, non-compliance with metabolic in water levels and then uploading data to the user’s smart-
treatment of nephrolithiasis is known to be as high as 70% phone via Bluetooth. Data are sent from the sensor to the
[5]. Recently, mobile health technology has gained atten- app each time the bottle detects a change in the water level
tion as a potential aide in helping improve compliance with and position through an accelerometer incorporated in the
medically indicated lifestyle and dietary treatments. Early sensor. Volume measurements are reset every 24 h and users
studies using “smart technology” and mobile health applica- are able to track their progress by accessing the application
tions have shown implementation of such strategies can not on their smartphone (Fig. 2).
only be beneficial in improving compliance, but also capable To assess the accuracy of fluid tracking in a real life
of facilitating sustainable behavioral change [6]. To date, fashion, eight healthy volunteers agreed to participate in a
studies looking at mobile health technology as it applies to preclinical trial. All study participants were given a Hidrate
increasing fluid intake are sparse, particularly among stone Spark™ water bottle and downloaded the associated soft-
formers. ware application on their smartphone using either Apple
Recently, a novel “smart” water bottle called Hidrate iOS™ or Android™. All participants underwent a step-by-
Spark™ (Minneapolis, MN) was developed and brought to step hands-on demonstration on how to appropriately set
market. This bottle is designed to be used as a non-inva- up and use the device, following the written instructions
sive fluid intake monitoring system, allowing users to track provided by the manufacturer. To ensure baseline accuracy
daily fluid intake in real time via smartphone to help achieve in recordings, volunteers were asked to fill the bottles com-
hydration goals. Such a device may have considerable poten- pletely to 24 ozs. and place it on a flat surface for 10 s to
tial as a meaningful tool in the care of patients with stone allow calibration. They were then asked to drain all fluid
disease as achieving high fluid intake is known to be dif- from the bottle and ensure that an accurate measurement
ficult. We sought to assess the accuracy of this device in a had been recorded to the application. All users were taught
preclinical setting prior to applying it towards patient care. how to calibrate and recalibrate the bottle to ensure consist-
ent readings. Users were given permission to recalibrate the
bottles on their own in the event of perceived inaccuracies
in fluid tracking.
Participants were then asked to use the bottle over a
2-week period whereby they were asked to track the precise
Fig. 1 a Hidrate Spark™ smart water bottle, b sensor stick
13