Modeling the cost-effectiveness of replacing conventional in-person follow-up care with mobile (smartphone) follow-up care in an ambulatory surgery patient population.

1. REPLACING AMBULATORY SURGICAL FOLLOW-UP VISITS WITH SMARTPHONE HOME MONITORING:
MODELING COST-EFFECTIVE SCENARIOS
Kathleen Armstrong1,2, Peter Coyte2, John Semple1,3
1. Division of Plastic & Reconstructive Surgery; 2. Institute of Health Policy, Management, and Evaluation, University of Toronto; 3. Women’s College Hospital
Introduction
Technology is identified as an opportunity to constrain the growth in healthcare costs
and eliminate barriers due to distance [1]. Ambulatory surgery offers a large and
growing patient population that could benefit from technology. Women’s College
Hospital (WCH) has completed a feasibility study using a mobile application (QoC
Health Inc, Toronto) to support postoperative care in breast reconstruction patients
(Figure 1). This feasibility study suggests that mobile follow-up care adequately detects
postoperative complications and eliminates the need for in-person follow-up care.
This is concordant with other postoperative telemedicine studies that use telephone
follow-up amongst various ambulatory surgery populations [2–4].
Figure 1: QoC Health Inc Mobile Application
Objective
This study models the cost-effectiveness of replacing conventional, in-person
postoperative follow-up care with mobile follow-up care following ambulatory breast
reconstruction. A societal and narrower healthcare system perspective are presented.
Methods
TreeAge Software is used to model cost-effective scenarios in which mobile follow-up
care replaces in-person follow-up visits over the first postoperative month.
Healthcare System Borne Costs (Table 1): In-person follow-up costs were
obtained from WCH and OHIP. In keeping with cost-effectiveness analysis, non-
recoupable or sunk costs are not included [5]. Smartphone follow-up costs were
provided by QoC Health Inc. Start-up fixed costs were divided over the number of
patients served over the useful lifespan of the technology (≈ 5 years). A surgical e-
assessment billing fee was assigned based on actual OHIP telemedicine fees. The
variable mobile costs are equivalent to $3.50 per patient per day.
Patient Borne Costs were calculated using patient demographic information. In-
person follow-up costs include foregone leisure costs, which are based on labour force
participation rates and age-sex adjusted average Ontario wages. We presumed that a
caregiver equivalent would be present at the first follow-up visit, and assigned a
homemaker wage ($11.28/hour) to that person [6]. The hourly rates were multiplied
by the travel time and length of the clinic visit. Travel costs were based on distance
from home postal code to WCH and Canadian Automobile Association (CAA)
Ontario-based average costs. Smartphone follow-up costs were modeled based on a
bring-your-own-device (BYOD) format. Each submission takes ≈3 minutes to enter
and uses 0.35 MB of data. In the feasibility study, patients submitted once daily for
the first two weeks and then once weekly for the next 2 weeks. Leisure time is not
interrupted.
In Ontario, 100 MB of data can be purchased for $10; therefore, data costs are
negligible. Patient training requires no additional time commitment.
The effectiveness of mobile and in-person follow-up care is measured as successful
surgical outcomes at 30-days postoperative. Prospective trials from the literature in
comparable ambulatory surgery patients were used to model effect [2-4].
Table 2 demonstrates how an INB only favours in-person follow-up if a 6 percentage
point difference in effect exists between the two follow-up groups and the patient
makes < $19.00/hr (highlighted in green). This calculation uses a willingness to pay
of $100,000 USD ($109,970 CAD based on the current exchange rate) per quality
adjusted life year (QALY), and a 0.04 QALY difference between no complication and
minor skin infection, based on the literature [7].
Table 1: Follow-up costs over the first 30 days postoperative
In-person
Mobile (enroll 1000 patients/yr)
Healthcare System Costs
Healthcare System Costs
Fixed Costs
Fixed Costs
Compensation
63.25
Health Centre Setup
1.39
Equipment
1.31
Design/Setup Procedure Prot
6.94
Variable Costs
Training
0.44
Drugs
0.13
Variable Costs
Other (Linens)
2.33
Platform Licensing, Accounts
42.00
Clinical assistant
6.25
Standard Support
43.05
Surgeon Fee
26.55
Infrastructure Hosting
19.95
Resident
6.44
Surgeon Fee
22.00
Subtotal H.C. per 30D
$174
Subtotal H.C. per 30D
$136
Patient Costs
Patient Costs
Variable Costs
Variable Costs
Patient Leisure Time
62.34
Patient Leisure Time
0
Caregiver Wage
33.84
Data (approx. 350 kB per
0
Travel (To and from clinic)
23.24
transmission with photo)
Parking
20.00
Subtotal P.C. per 30D
$207
Subtotal P.C. per 30D
$0
Grand Total (per 1.64 visits)
$381
Grand Total (per patient per 30D)
$136
Results
This modeling assumes that 1,000 patients are enrolled in BYOD mobile follow-up
per year and that in-person follow-up patients attend 1.64 visits within the first month
postoperatively. It also assumes an equivalent complication rate based on the
literature [2-4]. The results of this analysis are summarized in Table 1. From a
societal perspective, in-person follow-up care is $245 CAD more expensive than
mobile follow-up care. From a healthcare system perspective, in-person follow-up is
$38 CAD more expensive. The incremental cost-effectiveness ratio (ICER) is not
reportable. The incremental net benefit (INB) of mobile follow-up care
reflects the cost difference between the two interventions for any
willingness-to-pay value (societal INB=$245 CAD).
Scenario Analysis: Even when in-person follow-up is reduced to one visit per
patient over 30D postoperative, mobile follow-up is less costly from a societal
perspective. Mobile follow-up care remains cost equivalent to in-person follow-up
even when 100% of the mobile follow-up care patients attend one in-person visit.
Two-way Sensitivity Analysis: Patient wage was set between a homemaker and
average age-sex adjusted per hour wage. The mobile effect was varied between a
90-96% success rate.
Table 2: Two-way sensitivity analysis with varying patient lost leisure
time and effectiveness of mobile follow-up care
Patient Lost Leisure Time ($)
Delta Effect
$33.84
$56.98
$80.12
INB @
0.00
198.75
236.70
274.65
INB @
-0.02
110.77
148.72
186.67
INB @
-0.04
22.80
60.748
98.70
INB @
-0.06
-65.18
-27.23
10.72
Probabilistic Sensitivity Analysis (Figure 2): In all 100,000 simulations, mobile
follow-up care was cheaper than in-person follow-up care from a societal perspective.
The distributions assigned simulated 50% of scenarios where mobile follow-up was
less effective that in-person follow-up care. Scenarios that are less effective and less
costly, can still be considered cost-effective.
Note: WTP = $4,398.80 CAD per effect based on $109,970 CAD per QALY and
0.04 QALY assigned to one superficial skin infection [8].
The Bottom Line
Smartphone follow-up care is suitably targeted to low-risk postoperative
ambulatory patients. It can be cost-effective from a societal (INB = $245
CAD) and healthcare system perspective (INB = $38 CAD). It remains
cost-effective from a societal perspective even when compared to minimal
in-person follow-up or even if all patients from the mobile group were to
attend one in-person follow-up. Smartphone follow-up care has the
potential to generate huge cost savings when applied to a large and
growing ambulatory surgery patient population.
Figure 2: Probabilistic sensitivity analysis of mobile vs in-person
follow-up care
Less costly,
more effective
Less costly,
less effective
COI: Dr. Semple is a shareholder in QoC Health Inc. WCH REB and COI management plan approved. Funded by the CIHR e-health catalyst grant.
Breast
Reconstruc-on
Breast
Reconstruc-on
Breast
Reconstruc-on