Unforeseen Hurdles in Digital Health: Lessons from Medical Pros

One Unexpected Challenge I Faced Implementing a Digital Health Solution in My Practice

As owner of Superior Physical Therapy in Traverse City, Michigan, and a practicing PT for over 15 years, I’m always looking for ways to improve patient outcomes while keeping our independent clinic strong. A few years ago we launched remote therapeutic monitoring (RTM) using our own MyMovementRx platform — custom home exercise programs, automated tracking, and remote oversight — to boost adherence and open a new revenue stream via RTM CPT codes.

I expected technology challenges. What I didn’t expect was the hidden strain on our team’s capacity.

Even with excellent software (drag-and-drop HEP builder, 5,000+ exercise videos, and automatic CPT time tracking), the daily work of reviewing patient logs, sending check-ins, documenting time, and flagging non-adherent patients quickly added up to several extra hours per week per therapist and admin. Our clinic was already at full capacity with 1:1 care and documentation. The monitoring workload started pulling staff from direct patient time or forcing after-hours work. Early patient engagement suffered because follow-ups weren’t consistent enough.

Like many clinic owners, I assumed we could “just handle it in-house” once we had the right software. Patient compliance is the issue everyone talks about — but our real bottleneck was internal bandwidth. True remote monitoring is far more labor-intensive than most realize if you want high adherence and reliable billing.

We solved it by adopting a hybrid full-service model. Our on-site clinicians still prescribe programs and make clinical decisions, but we built an in-house virtual monitoring team of licensed PTAs and COTAs to handle routine check-ins, documentation, risk escalation (via our patent-pending stratification), and billing reports. This freed our staff, dramatically improved consistency, and boosted results: ~40% higher home exercise adherence and strong functional gains across more than 2,200 patients.

The virtual team made scaling simple and turned RTM into a reliable profit center instead of a burden.

Implementing RTM has been one of the best moves we’ve made for patient care and profitability — once we addressed the staffing reality. Don’t fall for the “set it and forget it” myth. Plan for the human element and the results will follow.

The unexpected challenge wasn’t technological — it was that the digital health tool I rolled out worked exactly as advertised, and our patients started receiving more information than they could absorb.

We integrated a comprehensive symptom-tracking and biometric-data platform into our concierge practice about two years in. Patients could log symptoms, sleep, energy, mood, weight, and connect their wearables. The dashboard was beautiful. The integration was clean. The data was rich.

What we hadn’t anticipated: patients now had access to a flood of their own physiological data without the framework to interpret what mattered. A woman would log three days of slightly elevated resting heart rate and arrive at her next visit anxious about her cardiovascular health. A woman whose sleep tracker reported one bad night out of seven would come in convinced something was wrong. The tool we’d added to support their understanding was, in some cases, raising their anxiety faster than it was answering their questions.

How we overcame it: we built a one-page interpretive layer between the data and the patient. Every dashboard a patient looks at now has a brief note explaining what counts as a meaningful signal versus normal day-to-day variation, with a clear “when to call us” trigger built in. We also added a brief weekly clinician note responding to anything the patient logged that might worry them, before they have a chance to worry alone.

The advice I’d give to anyone implementing digital health tools: the data itself doesn’t deliver value. The interpretation does. Build the interpretation layer before, or alongside, the data layer. Otherwise the technology will produce anxiety faster than it produces insight.

One unexpected challenge was realizing that a “digital health” tool can quietly shift the focus of a visit away from the patient and toward the screen. In psychiatry, the conversation is the core of care, and pop ups, rigid templates, and constant box checking can break rapport and slow the work. I addressed it by being disciplined about workflow, using technology to organize information after the human conversation rather than during it. When the system tried to lead the visit, I pushed it back into the background so it functioned more like a silent assistant than a second patient in the room. My advice is to measure any solution by a simple standard: does it give clinicians time back, or does it demand more attention from them. Before rolling anything out widely, have the product team shadow a clinician for a full week to see where the real friction shows up. If the tool cannot fit the natural flow of care, it is not ready for a clinical setting.

The unexpected challenge in my dermatology practice was that patients consistently bypassed the secure portal photo-submission flow and emailed photos to staff personal email accounts instead, even though the portal was HIPAA-compliant. The portal flow took five taps and required login. Email took zero taps. PHI was bleeding out at the workflow level, not the policy level.

The fix was a 20-second video tutorial sent in the appointment confirmation text the day before each visit, plus a simplified one-tap portal-photo button on the patient’s home screen. Once the portal flow was as fast as email, the email-around behavior stopped within two weeks. The advice I would give other practices is to compete with the unsecure shortcut, not just block it. If your HIPAA-compliant tool is slower than the workaround, your patients will find the workaround, and you will be the one defending it in a breach review.

The unexpected challenge wasn’t the technology, it was how quickly clinical staff stopped trusting it once early friction appeared. At our private addiction treatment clinic near Warsaw, we rolled out a multilingual patient-facing intake and content infrastructure to serve patients in eleven languages, and the technical build went smoothly. What we underestimated was how a single bad translation, one mismatched form field across language versions, or one delayed sync between systems would erode staff confidence in the entire stack. Within two weeks, intake coordinators were quietly reverting to manual workflows because they didn’t trust what the system was telling them.

What we did to recover: first, we rebuilt the rollout around what I now call the “trust ledger.” Every digital workflow had to demonstrate three weeks of zero discrepancy with the manual equivalent before staff were asked to abandon the manual fallback. That meant running parallel systems longer than felt efficient, but it preserved clinical confidence, which is the actual currency of digital health adoption.

Second, we appointed a clinical-side owner for each digital workflow, not a technical owner. The medical director (myself) and head of intake had named accountability for the patient-facing systems, and only then did the IT and content teams build to clinical specifications rather than assumptions. This single change reduced post-launch friction more than any technical fix.

Third, we built failure visibility into the system. Instead of hiding errors, every misalignment surfaced as a flagged item in a daily review queue. Staff stopped feeling like the system was “broken in mysterious ways” and started seeing it as something they could correct.

My advice to anyone implementing digital health solutions: the technology will work. The bottleneck is always clinical trust. Plan for the trust-building phase as a real, named, resourced project. Skip it and you’ll lose adoption permanently, regardless of how good the underlying tool is.

Andrzej Kulesza, Co-Founder & Medical Director, Zeus Detox & Rehab

One thing that caught me off guard wasn’t the tech, it was people.

When we started rolling out a digital workflow inside our operations, I assumed the biggest challenge would be integration or bugs. That part was actually manageable. APIs break, you fix them. Data sync issues show up, you debug and move on. What I didn’t expect was how resistant smart, capable people would be to something that clearly made their work easier.

We had a system that cut manual effort by almost 40 percent. On paper it was a no brainer. But the team kept going back to old habits. Copy pasting data, double checking things the system already handled, even avoiding the tool completely in some cases. It wasn’t logical, it was comfort. People trust what they’ve done for years, even if it’s slower.

I handled it the wrong way at first. Tried to push adoption hard, more training, more explanations, more “this will save you time.” Didn’t work. What finally moved things was sitting with the team, watching how they actually worked, and adjusting the system to fit their flow instead of forcing them into mine. Small changes made a big difference. Also, showing real examples helped. Not features, but actual before and after results tied to their daily tasks.

The biggest shift came when a few early adopters started seeing wins and talking about it themselves. That peer effect did more than anything I said.

If I had to give advice, don’t assume the challenge is technical. It’s usually behavioral. Build something simple first, let people get small wins, and don’t try to flip everything overnight. Also, stay close to the ground. If you’re not watching how people use the system in real situations, you’re guessing. And guessing is expensive.

I’m Dr. John Hegazin, and I run Bradenton Implants & Smile Center in Bradenton, where we use modern technology every day for implant, restorative, and general dental care. One unexpected challenge with digital tools wasn’t the hardware–it was getting the digital record to actually improve decisions instead of just creating more screens and clicks.

A good example was treatment planning for bigger restorative and implant cases. Digital imaging gives you a lot of information fast, but if the workflow isn’t tight, patients can feel overwhelmed and the team can lose time translating images into a clear, step-by-step plan. We fixed that by standardizing how we review images, explain options, and connect the scan directly to a personalized treatment plan patients can understand.

I learned that convenience alone doesn’t make a digital solution successful. It has to reduce confusion for both the team and the patient, especially when you’re discussing choices like a filling vs. crown vs. implant or mapping out full-mouth work.

My advice: start with one problem, not one piece of tech. Build a simple workflow around it, train the team on the exact words they’ll use with patients, and make sure the digital tool shortens the path to clarity, comfort, and follow-through.

One challenge we ran into was patient adoption of our online booking and patient portal system. Many of our patients, particularly older ones managing chronic pain conditions, weren’t comfortable navigating digital platforms and would still default to calling the office for everything, which largely defeated the purpose of implementing the system in the first place. There was also a language barrier component: since our practice serves patients who speak Arabic, Korean, and Spanish, navigating an English-only digital interface added another layer of friction.

What helped most was having our staff walk patients through the portal one-on-one during their visit rather than simply handing them a flyer or sending a generic email. We also made sure our multilingual team members were involved in those conversations so patients felt supported in their own language. Over time, that hands-on, personalized onboarding made a real difference in adoption rates and reduced the call volume on our front desk considerably.

My advice to other practices: don’t assume a digital solution sells itself. Technology only improves efficiency if your patients can actually use it confidently, and that requires investing just as much thought into the human side of the rollout as the technical side.

As the CEO of ProMD Health and a former Johns Hopkins researcher, I’ve overseen the integration of advanced technology into high-growth clinical environments. A major hurdle we faced was managing patient expectations when implementing AI simulation tools across our multi-location medical facilities.

While the tech provided a visual roadmap, it unexpectedly created “digital dysmorphia,” where patients expected a literal replica of the AI image rather than a clinical outcome. I drew on my experience as a firefighter and EMT to apply crisis-leadership tactics, refocusing our staff on grounding these high-tech simulations in medical reality.

We utilized Crisalix 3D AI to bridge this gap, treating the software as an educational tool rather than a final promise. This approach ensured our ethics-first culture, which earned us the BBB Torch Award, remained intact while we scaled our digital offerings.

For those implementing similar tools, don’t just train your team on the software; train them on the “psychology of the preview.” Ensure your digital solutions serve to enhance the human consultation rather than dictating the medical outcome.

One unexpected challenge was assuming that patients would naturally adapt to digital tools like online forms, patient portals, and digital communication. What we found is that access is not the same as comfort. Some patients, especially older adults or families with language barriers, struggled with the transition.

We overcame it by slowing down the process. Instead of forcing everything digital, we walked patients through it in the office, kept printed options available, and made sure our team could support both digital and non-digital workflows. That balance made a big difference.

My advice is simple: don’t implement technology just because it’s efficient for the practice. Make sure it works for your patients. If they can’t use it comfortably, it becomes a barrier instead of a solution. The goal is better care, not just faster systems.

The biggest change I made was separating clinical work from administrative noise. In my own practice, the stress was not only from seeing patients. It was the constant interruption from scheduling problems, phone calls, paperwork, follow-ups, and small coordination issues that kept pulling attention away from the exam room.

That experience is part of why I built Medical Staff Relief. I saw that burnout often grows when every task lands on the same few people, with no clear structure around who owns what.

The change was creating clearer workflows and moving repeatable administrative work to trained support staff. Scheduling, intake, documentation support, billing coordination, and patient follow-up needed ownership, not random attention between patients.

The workday became less reactive. There was less scrambling, fewer interruptions, and more focus. A less stressful practice usually doesn’t come from asking people to work harder. It comes from designing the day so fewer things break.

The biggest surprise? ADHD brains tend to reject the very digital tools meant to help us. We’re wired for novelty, we get a hit of dopamine from downloading a new app, use it for a week or two, then forget it exists. I’ve watched clients rack up thirty, forty productivity apps on their phones and never stick with a single one.

I figured out the hard way that you have to start analog. Build the habit without tech, then, maybe, add it later. I have people use paper planners and those old kitchen timers first. Once they’ve got the routine down, we add one digital tool. Just one.

Here’s what I tell people: digital isn’t automatically better for ADHD. The fanciest app in the world is useless if it makes your life more complicated. Pick tools that make things easier, not harder. And keep something low-tech around for the days when all the apps and notifications are too much.

The most unexpected thing I ran into rolling out our AI receptionist into a multi-location dental group last year wasn’t technical — the integrations all worked, the voice quality was good, the booking handoff into their PMS was clean. The unexpected challenge was that the front desk staff at the practices started routing every difficult phone call to the AI on purpose, because the AI was easier to deal with than the patient. Insurance verification calls, angry billing calls, anything that wasn’t a friendly book-an-appointment call, they’d just hand it off and walk away.

Which on the surface sounds like adoption working. The numbers looked great. Call answer rate hit 100%, hold times disappeared, after-hours capture went through the roof. But what we noticed three months in is that patient satisfaction in the offices that overused the handoff actually dropped, even though appointment volume was up. The AI was excellent at scheduling, fine at billing questions, but not good at the human emotional layer of “I’m scared about this procedure” or “I just lost my insurance and I don’t know what to do.” Patients felt processed, not cared for, when they wanted the latter.

The fix took us about six weeks. We changed three things. First, we built a sentiment-detection layer into the conversation that flags any call where the patient sounds upset, anxious, or financially distressed, and force-routes it to a human regardless of staff preference. Second, we added a daily report to each office manager showing which call types the AI handled and which got escalated, so the office could see the pattern of avoidance instead of just the volume number. Third, and most important, we ran a workshop with the front desk teams on what the AI is for and what it isn’t — framed not as policy, but as “here’s where you’re irreplaceable, and here’s where you don’t have to be.”

The advice I give other practices implementing digital health: the technology rarely fails. The implementation fails because nobody plans for the social system around it. Build the metrics that show you when staff are using the tool to disengage, and treat that signal as seriously as you treat downtime.

I’ve spent decades implementing secure, scalable IT systems as the founder of Impress Computers, and the most unexpected challenge in digital health-style rollouts is usually not the software — it’s access chaos. The tool works, but people lose time waiting on permissions, juggling logins, or using unsafe workarounds because the “official” process is too slow.

I’ve seen the same pattern in other high-trust, high-compliance environments like law firms. When secure systems are added without clean access rules, staff start emailing files, saving things locally, or sharing credentials just to keep moving — which creates both risk and friction.

What fixed it was boring infrastructure work: map who needs access to what, set it up correctly on day one, use MFA, and remove the dependency on one person who “knows how it works.” In practice, that means treating onboarding, permissions, device setup, and backups as part of the solution — not as admin leftovers.

My advice: before rollout, ask your team three things — where they lose time, where they get stuck waiting, and which system makes their job harder than it should. If you solve those bottlenecks first, adoption gets easier and the digital health solution actually improves care instead of becoming one more layer of frustration.