|Year : 2018 | Volume
| Issue : 3 | Page : 133-141
An observational study to assess the feasibility of remote monitoring of patients in the early postoperative period after elective surgery
Omar Faiz1, Subramanian Nachiappan1, Chukwuemeka Anele2, Emma-Jane Roberts3, Chris Barker4
1 Department of Surgery and Cancer, St Mark's Hospital and Academic Institute, Harrow, UK
2 Department of Surgery and Cancer, St Mark's Hospital and Academic Institute, Harrow; Department of Cancer and Surgery, Imperial College, London, UK
3 Health and Life Sciences Department, De Montfort University; Spirit Health Group, Leicester, UK
4 Spirit Health Group, Leicester, UK
|Date of Web Publication||18-Oct-2018|
Spirit House, Saffron Way, Leicester, LE2 6UP
Source of Support: None, Conflict of Interest: None
Background and Objectives: The utility of postdischarge remote monitoring with patient self-measurement of physiological parameters has not been utilized in the elective colorectal surgical setting till date. Before full application of this utility, patient-acquired measurements' accuracy, and ability to detect deviation from normal physiological must be assessed. This study aims to ascertain the congruency of patient-measured and nurse-measured readings in the postoperative in-hospital setting, before future application in the home setting. Methods: This is a prospective single-institution study comparing patient and nurse measured readings of blood pressure, heart rate, oxygen saturation, and temperature, in ambulatory patients in the postoperative setting. Patients were provided with and trained on the usage of a handheld computing device with bluetooth-enabled measuring adjuncts preoperatively. Patients were instructed to acquire two sets of readings per day for 2 days. These were compared with nurse-acquired measurements and extrapolated to the modified “National Early Warning Score” system. Inter-rater concordance was analyzed using the Bland–Altman method for raw physiological measurements and for modified-National Early Warning Score (NEWS) risk and response categories. Results: A total of 40 elective colorectal patients were prospectively approached. Twenty-seven completed the study with 25 patients (15 males, age range 18–79, undergoing a variety of colorectal procedures) accruing a total of 98 datasets of physiological parameter readings. There were no significant differences/bias between blood pressure (P = 0.572), heart rate (0.122), and oxygen saturation (P = 0.308) readings. Patient and linked nurse measurements for all patients elicited the same modified NEWS clinical risk and required response in all instances. Conclusion: Patient self-measurement of physiological parameters and subsequent derived modified NEWS categories using a remote monitoring system are comparable to nurse-acquired measurements. Remote monitoring is technically feasible, and the next steps would be to trial this technology in the home setting.
Keywords: Colorectal surgery, digital health, digital medicine, elective care, health, mobile technology, remote monitoring
|How to cite this article:|
Faiz O, Nachiappan S, Anele C, Roberts EJ, Barker C. An observational study to assess the feasibility of remote monitoring of patients in the early postoperative period after elective surgery. Digit Med 2018;4:133-41
|How to cite this URL:|
Faiz O, Nachiappan S, Anele C, Roberts EJ, Barker C. An observational study to assess the feasibility of remote monitoring of patients in the early postoperative period after elective surgery. Digit Med [serial online] 2018 [cited 2019 May 20];4:133-41. Available from: http://www.digitmedicine.com/text.asp?2018/4/3/133/243641
| Introduction|| |
The health system has been tasked with providing innovative integrated models of care, transforming services to improve patient experience, outcomes, and efficiency of care. A key focus of health-care systems globally includes better integration and joined up working across sectors and the harnessing of technology within health care to simplify patient access to care, in the most appropriate location while supporting people in managing their own health. It has been widely documented that directly involving patients in managing their condition empowers patients and facilitates a better use of health-care resources.
The simple use of mobile technology can empower patients and clinicians with tools to support the safe, proactive, and efficient management of health. Remote monitoring solutions, delivered over intuitive tablet devices, provide a powerful platform for intensive monitoring, education, and empowerment of at-risk patients. These solutions are custom-designed to help patients engage with their health and self-manage their conditions through the daily capture of physiological data and self-reported information about their wellbeing. That data are linked to dynamic care plans and evidence-based algorithms that enable automated triage for healthcare teams when an escalation of care is required. Crucially, this gives clinicians' real-time visibility of physiological trends to help them recognise and respond to deterioration. Moreover, it provides them with an evidence-base “remote control” of vulnerable patients, allowing them to intervene proactively rather than wait for a costly emergency exacerbation. This is hugely reassuring for patients and carers, who also feel in greater control of their care.
Adoption of remote monitoring solutions is growing across all settings within healthcare globally. Evidence shows its helping to facilitate earlier discharge, prevent readmission and reduce the risk of future exacerbations through better self-management. Digital healthcare solutions can provide the right care at the right time. Moreover its why remote monitoring solutions, which give clinicians unprecedented real-time visibility of their most at-risk patients, must play a key role in future models of health care.
Colorectal surgery has been evolving rapidly with the transition from open conventional surgery to minimal access surgery with the aid of technological innovations and advances. Together with preoperative stratification of patients, improved operative techniques and good postoperative management, clinicians are able to send patients home earlier. For instance, laparoscopy was found to improve short-term outcomes postoperatively and when several aspects of perioperative care were formalized within an enhanced recovery programme protocol, better outcomes were also noted., Concurrent with the changes in colorectal cancer surgery, health-care costs have also been increasing in general. There has been considerable strain on hospital resources and high bed occupancies in most hospitals in the United Kingdom National Health Service (NHS), as well as internationally.,,
One of the methods to tackle this would be to enable patients to be safely discharged sooner after elective surgery, thereby increasing bed capacity and increasing patient turnover. This however must not impact on patient care and must not result in increasing readmissions soon after elective surgery and early postoperative discharge. Specifically in colorectal surgery, Schnieder et al., in their epidemiological study noted that readmission rates after colectomies have increased during the past 20 years with the mean length of postoperative stay declining. A similar trend was also noted by Hendren et al., who noted a higher risk-adjusted readmission rate in hospitals where elective colorectal resection patients were discharged a median 4 days or less after surgery compared to a median of 5 days. This is an interesting dilemma faced by surgeons in terms of safely discharging patients in a timely fashion and minimising readmission rates with early postoperative complications. In fact, Lawson et al., noted that postoperative complications after colectomy are predictive of readmission and opined that preventing and proactively managing post-operative complications may be the most important step toward reducing readmission rates.
In our institution, elective surgical patients are discharged home when they are safely and independently mobilizing, their physiological parameters and clinical examinations are satisfactory and blood markers are normalizing. The patients are subsequently reviewed in the outpatient clinic in approximately 2 weeks, to ensure good progress in postoperative recovery and also to discuss and put in place further follow-up and management. There is also a telephonic service in place currently whereby a dedicated contact point is maintained by two experienced colorectal nurse specialists who were based on the dedicated colorectal ward. The discharged patients can call in for regarding any issues and this utility functions to allay concerns of the patients and if need be, allow them to organise a clinical review if necessary. This “patient contact” is documented in the patient's electronic notes for future reference as well. This is a patient-initiated process, which is often instigated when the patient subjectively feels unwell or has any concerns. There is no objective assessment that may indicate that the patient is unwell. There are several reasons why a colorectal resection patient may feel unwell in the early postoperative period. The first postoperative month is when patients are most at risk of complications. These include anastomotic leaks, bleeding, surgical site infections, chest infections, urine infections, dehydration, and cardiovascular and cardiorespiratory complications such as myocardial infarctions and pulmonary embolisms.,,
In the current climate of promoting safe early discharge, it is vitally important that postoperative complications are not missed. In fact, it would be optimal if patients could be discharged early with a mechanism to detect possible complications and ideally detected them early. This will first allow quicker and more proactive rather than reactive management of the complication. Second and more importantly, it would most likely result in better patient outcomes as the complication would be detected and managed earlier on rather than when the patient is severely unwell secondary to the complication and sustains significant morbidity.
One of the first objective indicators of a patient becoming unwell would his or her physiological parameters deviating from the normal. Therefore, a good method of detecting a possible complication would be to have regular observation and documentation of physiological observations. This is routinely done on the surgical ward in the early postoperative period. This routine measurement can also be remotely obtained from the patient's home with the advent of mobile tablet PC and associated measurement devices already available in the market. Currently, patients with chronic conditions such as chronic obstructive pulmonary disease, asthma, diabetes, and rheumatoid arthritis are monitored at home using tablet computers with Bluetooth-enabled physiological measurement devices, linked to secure servers accessible by clinicians.,, Others have also trialed remote monitoring in the post-operative period in other surgical specialties such as cardiac surgery and hip and knee surgery.,
In the current pilot study, the remote monitoring system would be trialed in the hospital setting in patients undergoing elective colorectal surgery. The primary aim is to assess the congruency of patient-acquired physiological measurements with contemporaneous nurse-acquired measurements, and whether the two elicit the same clinical response according to established patient monitoring algorithms within the hospital setting. This pilot study will subsequently be followed up with further studies assessing this utility in the post-discharge home setting.
| Methods|| |
Variables and definitions
This was a prospective study in a single institution. Consecutive patients aged 18 and above, undergoing elective colorectal resectional surgery, were approached preoperatively in the outpatient clinic. Patients were included if they were expected to comply with the study procedures and were competent in using a handheld computer/device or were agreeable to learning how to.
Patients undergoing both laparoscopic and open procedure were included. Patients suffering a recurrent cancer were not included and those who were due a planned admission to intensive care unit postprocedure were not included. Those who were due to be electively admitted to high-dependency unit (HDU) in the immediate postoperative period/days, and subsequently transferred to a normal colorectal ward were included.
Training on the device was administered at preoperative clinic and again when patients received it postoperatively. Patients also received literature on the handheld device in the clinics. Patients were informed that the devices will be given to them on the ward on the postoperative day (POD) 2 or when they were able to mobilise on the ward and felt physically able to take part in the study, whichever occurred later.
Patients were free to decline participation at any point in the study. Patients who met the inclusion and exclusion criteria, were subsequently recruited after informed written consent.
The handheld device is a tablet device [Figure 1] with a proprietary software system owned by Spirit Healthcare Limited (Spirit Healthcare). This software and device, called CliniTouch, is used to remotely monitor patients' physiological markers in chronic conditions such diabetes and chronic obstructive pulmonary disease within the NHS setting. The handheld tablet is CE-marked as are the associated sensor devices (blood pressure machine and pulse oximeter), both of which are linked to the former via Bluetooth technology.
|Figure 1: CliniTouch handheld device, Bluetooth-enabled sphygmomanometer and pulse oximeter|
Click here to view
The system would as, previously done in the chronic conditions, allow the measurement of pulse rate, blood pressure and oxygen saturations, and in addition, temperature will be manually inputted by the patient after measurement with a digital thermometer. The measurements would be transmitted to a secure online server, accessible via a hub computer available to the clinician/researcher in-hospital, to be reviewed on a daily basis.
The system also has video conferencing and high definition still image transferring capabilities over a secure 3G mobile wireless connection to the clinician's hub computer in the surgical office/ward. This utility is available for direct video conferencing with the patient. For the purposes of this study, the video conferencing capability and image transfer function were not utilized. Future studies will investigate the utility of remote monitoring and communication with the hospital/clinicians in the postoperative period, with the patient having been discharged home.
Patient demographics such as age, sex, operative indication, primary operative procedure, operative access, admission duration (length of stay), and postoperative start day and end day of patient self-measurements were collated.
Each patient was instructed to take his or her physiological measurements twice a day: once between 0600 and 1500 h, and once between 1501 and 0000 h. Data were retrieved from the online server for analysis and comparison with nursing measurements taken on the ward.
Four data points were collated for each measurement event: systolic blood pressure (and diastolic which was not included in the analysis of NEWS scoring, see below), heart rate, O2 saturations, and temperature.
Nursing measurements of the above four were also collated. Nursing measurements of physiological parameters usually took place up to 6 times every 24 h depending on clinical need. Each patient self-measured reading was matched with a nursing reading, within the 2 time periods described above. All patient-measured readings and matched nurse-measured readings were within 2 h of each other. Each patient was requested to obtain readings for 48 h, therefore acquiring a total of four sets of readings.
Modified “National Early Warning Score”
The National Early Warning Score (NEWS) is a validated system aimed at standardising monitoring methods and documentation at the UK national level. It is specifically for patients in hospital and can be used in a prehospital setting as well. In this study, the NEWS system was used in postoperative patients in the in-hospital setting. The nursing documentation would be on an A4 sized paper and a clinical responses to a NEWS trigger range from 0 and 1–4 (low clinical risk), 5–6 or a single score of 3 in one parameter (medium clinical risk), and 7 or more (high clinical risk).
In the hospital setting, these scores would direct clinical responses ranging from continuing with routine observations, to urgent medical review and management [Table 1] and [Table 2]. For the purposes of this study, the NEWS score was modified such that, respiratory rate, presence of oxygen supplementation and mental alertness were not included in the aggregate NEWS score. This was named “modified NEWS.”
The study aims to assess the inter-rater (patient and nurse) concordance in the modified NEWS and whether the appropriate clinical response is elicited by both the nurse-measured modified NEWS and the patient-measured modified NEWS.
- Inter-rater concordance between patient-measured reading and matched nurse-measured reading was analyzed utilizing the Bland–Altman method of assessing inter-rater differences and bias
- Concordance of clinical risk and response categories, between patient-measured and nurse-measured readings is described in a cross-tabular format to ascertain if both patient- and nurse-measured “modified NEWS” elicit the same appropriate clinical responses, according to the NEWS Traffic light categorization. Modified NEWS of up to four will require 12 hourly observations, 5–6 will require 4–6 hourly observations and a nurse review, and >6 will necessitate 1 hourly observations and a medical review.
| Results|| |
A total of consecutive 40 patients were approached preoperatively. Two patients declined participation due to physical inability (one had severe rheumatoid arthritis in her hands and the other was registered blind– required large fonts to read). A further seven patients declined preoperatively as well. Most of these patients were not keen to take measurements by themselves. Two patients withdrew their participation postoperatively, just before the start of the study, as they felt too tired and two were not keen to proceed as they were due to be discharged the next day (POD 3).
A total of 27 patients started and completed 48 h of self-measurements or up to their discharge. Data were collated from the hub computer for these 27 patients. Two of the patient's data were unreadable due to a technical fault on the handheld device. A further two patients obtained only three sets of measurements each as they were discharged on the second afternoon of the 48 h period.
In total, 25 patients' raw data (systolic blood pressure, heart rate, oxygen saturation, and temperature was accrued and matched nurse-measurements were collated from ward charts. There were a total of 98 sets of readings for analysis. Patient accrual is described in [Table 3] and demographics in [Table 4].
There were 15 males and 10 females. Eleven patients were aged between 18 and 50 years, ten between 51 and 70 years and 4 aged between 71 and 80 years. Operative indications included adhesions, cancer, inflammatory bowel disease, and polyposis syndromes. Two patients had an American Society of Anaesthesiologist Class I, and the rest were Class II. Six patients underwent open procedures and the remaining 19 were laparoscopically operated on. The operative procedures and procedure durations are described in [Table 4] together with the rest of the patient demographics and perioperative factors.
Twelve patients commenced on the study ( first self-measurement) within 3 days of their operation, 10 between POD 4 and POD 6, and the remaining three after POD 6. Five patients stayed in hospital for up to 5 days and 17 patients stayed between 5 and 10 days. Three patients, including two who spent the initial postoperative period in HDU stayed in hospital >10 days (range 8–21 days).
Congruency between patient-acquired measurements and nurse-acquired measurements was assessed utilizing the Bland–Altman method (see below). There were no significant differences/bias between nurse and patient acquired systolic blood pressure [Figure 2], P = 0.572], heart rate [Figure 3], P = 0.122] and oxygen saturation [Figure 4], P = 0.308] readings. There was significant difference in the temperature readings [Figure 5], P < 0.001]. The patient-acquired readings were significantly slightly lower than would be expected compared to the nurse-acquired readings by 0.5°C.
No significant bias or differences noted between patient-measured readings and nurse-measured readings (P = 0.572).
No significant bias or differences noted between patient-measured readings and nurse-measured readings (P = 0.122).
No significant bias or differences noted between patient-measured readings and nurse-measured readings (P = 0.308).
The patient-measured temperature readings were on average significantly lower than the nurse-measured readings by 0.5°C (P < 0.001).
Assessment of the clinical risk and responses showed that each patient acquired modified NEWS score would elicit the same response as the patient acquired modified NEWS score it is linked to, in all instances [Table 5].
| Discussion|| |
This pilot study has shown that there is no significant bias or difference between patient or nurse measurements with respect to blood pressure, heart rate, and oxygen saturations. Although we did not find any significant differences between the two in most of the parameters, we noted that the patient-acquired temperatures were slightly but significantly lower (0.5°C) compared to the nursing measurements. This may be due to a calibration issue with the digital thermometers supplied to the patients, or it may be due to patients not waiting for stabilization of the reading (evidenced by a “beep” tone), which may take up to 2 min. Nevertheless, overall, the study showed that modified NEWS scores derived from patient-acquired and matched nurse-acquired measurements fell within the same clinical risk categories and would elicit the same clinical responses in all instances.
Our study, therefore, showed that the utility of remote patient self-monitoring in the postoperative period is feasible and comparable to nurse-led monitoring and can be extended to and investigated in the postdischarge home setting in such patients.
This study investigated remote postoperative monitoring, which is similar to several studies in different specialties, in the postoperative period.,, Till date, there have been no studies in the literature that assessed the utility of remote monitoring of patient-measured physiological parameters in the postoperative period in colorectal surgery. However, Dawes et al. utilized a hand-held tablet device to remotely survey patients in the early postoperative home setting prior to the first postoperative clinic review, on specific qualitative domains such as diet, bowel function, fatigue, pain, ambulation, wound care, and general well-being. Although there were no quantitative measurement of physiological parameters, the authors noted that patients reported the device to be useful and allowed them to be more involved in their recovery.
The availability of quantitative measurements of patient physiological parameters, deviations of which are often the first indication of postoperative complications, we opine will not only be useful for the patient in a similar fashion but also aid the clinician in remotely monitoring the patient in the home setting and allow a proactive early management strategy if a postoperative complication were to occur. Furthermore, our remote monitoring system has in place the option to include daily surgery- and patient-specific survey, direct video communication and photo transfer utilities, which can be used by the clinicians to monitor and communicate with the patient from the surgical office or hospital. We opine that these utilities within the handheld device will yield similar views from patients. These added functions will be investigated and assessed in future studies in our institution.
There are several limitations in our study. First, the patient-acquired measurements and matched nurse-acquired measurements were obtained up to 2 h apart. Differing readings may result due to patient activity, pain and other factors, which may influence his or her physiological parameters. Nevertheless, we did not note significant differences except in temperature readings, which may be secondary to calibration issues or incorrect acquisition of the reading as described above.
A further limitation in the study is the omission of specific parameters utilized in the NEWS system in hospital. We elected to score “zero” for respiratory rate, oxygen supplementation and level of consciousness [Figure 1] and [Figure 2]. There is technology available that assessed respiratory rate in the market, but we did not have the option or the feasibility of having such a measurement device in our study. We also felt that a patient manually counting their breaths will not be an accurate assessment. The omission of respiratory rate is a significant limitation. However, we believe that the availability of all the other parameters is valuable nonetheless in monitoring a patient in the postoperative setting. We also elected to assume a patient in the home setting would not be mentally obtunded or need oxygen supplementation. Furthermore, oxygen saturation is an indirect marker of respiratory rate, and we noted no significant differences in nursing and patient self-measurements in this parameter.
Third, there were a variety of indications for surgery and subsequently types of surgical procedures undertaken in these patients. Patients undergoing major resectional surgery may have a slower recovery and may be less inclined to undertake self-monitoring compared to those who underwent smaller procedures such as closure of ileostomy. However, we specifically enrolled patients in the study only when they were in the later part of their postoperative hospital stay, to imitate as much as possible the patient's home setting where they are more ambulatory and independent. Furthermore, the variety of patients and operative indications in the study, lends itself well to the generalizability of the results to all colorectal surgical patients. Fourth, we may have a selection bias of patients who are more motivated and inclined to take part in the study. We noted that 13 out of 40 patients initially approached did not eventually take part in the data accrual for varying reasons.
Furthermore, we only required four sets of readings per patient for the purposes of this study. Although many of the patients were happy to continue with more self-measurement readings, the compliance rate of prolonged self-measurement has not been assessed in this study. Compliance may be further reduced in the home setting, when patients may be less inclined to take self-measurements, especially when they feel well. Conversely, the opposite may apply in patients who feel unwell and therefore may be more inclined to take their measurements. This needs to be further assessed in studies, being planned for in our institution, investigating our hand-held device and postoperative monitoring utility in the early post-discharge period.
| Conclusion|| |
This study has shown that the patient self-measurement of physiological parameters and subsequent derived modified NEWS categories using a remote monitoring system are comparable to nurse-acquired measurements. It has shown that remote monitoring is technically feasible and the next steps would be to trial this technology in the home setting of patients who have been discharged home after colorectal surgery. The options of video conferencing, photo transfer (e.g., of wounds, stoma, etc.) and online daily questionnaires to ascertain qualitative recovery parameters or quantitative parameters such as stoma output are also available in our remote monitoring system and will be assessed in future studies.
Financial support and sponsorship
Conflicts of interest
Barker and Roberts have an interest in Spirit Digital and CliniTouch Vie. All data was provided by St Marks Hospital, analysis of the data was reviewed by Nachiappan and final editorial control was given to Anele who is conducting phase two of the trial in the home setting.
| References|| |
Hasegawa H, Kabeshima Y, Watanabe M, Yamamoto S, Kitajima M. Randomized controlled trial of laparoscopic versus open colectomy for advanced colorectal cancer. Surg Endosc 2003;17:636-40.
Lawrence JK, Keller DS, Samia H, Ermlich B, Brady KM, Nobel T, et al.
Discharge within 24 to 72 hours of colorectal surgery is associated with low readmission rates when using enhanced recovery pathways. J Am Coll Surg 2013;216:390-4.
Adamina M, Kehlet H, Tomlinson GA, Senagore AJ, Delaney CP. Enhanced recovery pathways optimize health outcomes and resource utilization: A meta-analysis of randomized controlled trials in colorectal surgery. Surgery 2011;149:830-40.
Faiz O, Haji A, Bottle A, Clark SK, Darzi AW, Aylin P, et al.
Elective colonic surgery for cancer in the elderly: An investigation into postoperative mortality in english NHS hospitals between 1996 and 2007. Colorectal Dis 2011;13:779-85.
O'Dowd A. Hospitals are advised to halt elective surgery to free beds for winter. BMJ 2016;355:i6766.
Fusco M, Buja A, Piergentili P, Golfetto MT, Serafin G, Gallo S, et al.
Individual and hospital-related determinants of potentially inappropriate admissions emerging from administrative records. Health Policy 2016;120:1304-12.
Cvetkovic A, Cvetkovic D, Stojic V, Zdravkovic N. Length of hospital stay and bed occupancy rates in former yugoslav republics 1989-2015. Front Pharmacol 2016;7:417.
Qin S, Thompson C, Bogomolov T, Ward D, Hakendorf P. Hospital occupancy and discharge strategies: A simulation-based study. Intern Med J 2017;47:894-9.
Schneider EB, Hyder O, Brooke BS, Efron J, Cameron JL, Edil BH, et al.
Patient readmission and mortality after colorectal surgery for colon cancer: Impact of length of stay relative to other clinical factors. J Am Coll Surg 2012;214:390-8.
Hendren S, Morris AM, Zhang W, Dimick J. Early discharge and hospital readmission after colectomy for cancer. Dis Colon Rectum 2011;54:1362-7.
Lawson EH, Hall BL, Louie R, Zingmond DS, Ko CY. Identification of modifiable factors for reducing readmission after colectomy: A national analysis. Surgery 2014;155:754-66.
Dindo D, Demartines N, Clavien PA. Classification of surgical complications: A new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:205-13.
Kirchhoff P, Clavien PA, Hahnloser D. Complications in colorectal surgery: Risk factors and preventive strategies. Patient Saf Surg 2010;4:5.
Guinier D, Mantion GA, Alves A, Kwiatkowski F, Slim K, Panis Y, et al.
Risk factors of unplanned readmission after colorectal surgery: A prospective, multicenter study. Dis Colon Rectum 2007;50:1316-23.
Battista L. A new system for continuous and remote monitoring of patients receiving home mechanical ventilation. Rev Sci Instrum 2016;87:095105.
Steel S, Lock S, Johnson N, Martinez Y, Marquilles E, Bayford R, et al.
A feasibility study of remote monitoring of asthmatic patients. J Telemed Telecare 2002;8:290-6.
Health Quality Ontario. Home telemonitoring for type 2 diabetes: An evidence-based analysis. Ont Health Technol Assess Ser 2009;9:1-38.
McElroy I, Sareh S, Zhu A, Miranda G, Wu H, Nguyen M, et al.
Use of digital health kits to reduce readmission after cardiac surgery. J Surg Res 2016;204:1-7.
Toogood PA, Abdel MP, Spear JA, Cook SM, Cook DJ, Taunton MJ, et al.
The monitoring of activity at home after total hip arthroplasty. Bone Joint J 2016;98-B:1450-4.
Msayib Y, Gaydecki P, Callaghan M, Dale N, Ismail S. An intelligent remote monitoring system for total knee arthroplasty patients. J Med Syst 2017;41:90.
McGinley A, Pearse RM. A national early warning score for acutely ill patients. BMJ 2012;345:e5310.
Dawes AJ, Reardon S, Chen VL, Kaiser W, Russell MM, Ko CY, et al.
Wireless technology to track surgical patients after discharge: A pilot study. Am Surg 2015;81:1061-6.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]