Telemedicine may drive up medical utilization, increase cost for respiratory illness: RAND Health

click to enlargeIs convenience the culprit? Researchers from RAND Corporation’s Health program conducted a three-year study of telemedicine (here called telehealth) usage by employees of CalPERS for respiratory illness and came to a surprising conclusion. From the study abstract: “12 percent of direct-to-consumer telehealth visits replaced visits to other providers, and 88 percent represented new utilization. Net annual spending on acute respiratory illness increased $45 per telehealth user.”

The study examined 2011-2013 claims information for over 300,000 people insured through the California California Public Employees’ Retirement System, which despite the name provides health benefits to active state employees as well as retirees. It targeted common acute respiratory infections (sinus infections, bronchitis and related) to determine patterns of provider utilization and the change after the introduction of telehealth. Of that group, 981 used the Teladoc system for video consults, adopted by CalPERS in 2012.

The objective of the study was to determine whether the telehealth visits were new care or substituted for other types of care such as doctor, clinic, or ED visits. Even though the telehealth services were far cheaper–about 50 percent lower than a physician office visit and less than 5 percent the cost of a visit to the ED–they did not make up for the calculated 88 percent rise in utilization.

Similar results were reported by RAND in last year’s research on retail clinics, which estimated that 58 percent of visits for low-severity illnesses were new and not shifted from EDs or doctor’s offices. What is in common? Convenience. Convenience opens up greater use. If you have a store down the street, you may pop in daily versus once-weekly.

Updated: Some further insights from Mobihealthnews were that the study stated that telehealth visits may be more likely to result in additional costs, such as follow-up appointments, testing or prescriptions. In other words, the telehealth visit starts off less expensive, but the standard of care in follow-up adds to that initial cost.

The RAND recommendation is thus not a surprise: make more telemedicine visits a shift from office or ED to restrict telemedicine growth. Raise the cost of co-pays for the service to reduce demand. On the ‘high side’, encourage ED ‘frequent flyers’ to use telehealth services instead. Pass the painkillers. Health Affairs (abstract only; paid access required for full study), RAND Health press release.

Analysis: instead of self-doctoring, and suffering at home and in the workplace, the small group of CalPERS policyholders in the study actually used their new benefit to check their health–as intended! The additional cost is not staggering; doing some basic math, it is $45 x 981=$44,145 for the entire group.

But missing are two deductions to that cost: the cost for lost time off the job and the cost of spreading infection to co-workers. Those are equivalent to better outcomes, and, bless their hearts, PRODUCTIVITY.

Teladoc has already fired back, citing its Home Depot study with savings of more than $20 per enrollee per month. POLITICO Morning eHealth 7 Mar  Yet an earlier RAND study using the same group data, published in Health Affairs (February 2014) found that 34 percent of visits were on weekends and holidays, with 21 percent of visits were made by those not accessing care in the prior year, which directionally indicates higher utilization (#RISE2017 presentation).

Perhaps the wrong metrics are being assessed in these studies. In the rush to demonstrate ROI, to lower costs, there’s a certain myopia present. Technology is making healthcare more convenient to use. That argument was that if it is convenient, it’ll be better used and people will be healthier. Is that now a bad thing, even if 9 out of 10 acute respiratory conditions resolve themselves after time? (That time can be weeks for a nasty sinus infection not caught early.) Mobihealthnews points out that increase in utilization can be construed as a positive to undertreated conditions such as diabetes and mental health. Yet in thinking about these, these represent more complex and longer-term uses of telehealth that logically should add in vital signs monitoring and qualitative assessment, versus a bad cold which is visible by video and a few questions between sneezes. Undertreated conditions are more expensive, and less convenient, to monitor.

Low overall utilization of telemedicine is still the norm. Univera Healthcare, a western NY State insurer headquartered in Buffalo, found only 6 percent in their area have ever tried telemedicine, which is slightly surprising given the long distances and long winters in that area. But 80 percent of triers liked it. The online survey captured responses from 2,000 responders. Though NY State requires insurers to provide telemedicine in their policies, many reported that their insurers or doctors didn’t offer it, which is partly an educational problem. WBFO.org

Categories: Latest News.

Comments

  1. Donna CusanoDonna Cusano

    For our Readers: highly recommended. Plan to sit down with a cup, maybe two, of coffee or tea for this one. I’d recommend this analysis to anyone seriously interested in telemedicine cost. Mr Judson had accessed this full study and compared it to other research, including RAND’s in 2014. I could detect some of the faults but not to the extent he did. Thanks for this!

    This reminds me of the infamous and highly publicized Yale/Dr Sarwat Chaudhry telehealth ‘failure’ study of 2010 that monitored heart failure patients via IVR. However, the study required the patient to call in, without automated outbound calls which are standard for IVR programs, and did not use any other patient monitoring method. It was used to denigrate the effectiveness of telehealth for some time.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>