Patientory is a project of an American company, located in Atlanta (Georgia, USA). Their goal is to ensure a safe transaction of patient information data from one provider to another. A good goal, considering that most information regarding your health are very important for your everyday life – if your information gets public, the consequences can range from nonexistent to awkward to devastating, depending on your circumstances. If you don’t believe that such information can be absolutely ruinous, I really can recommend watching the movie “Philadelphia”. In this movie from 1994, the information of Andrew Beckett having AIDS led to his outing, as well as the inevitable downfall of his career and reputation in the US of the early 90s.
While the story of Philadelphia took place more than 20 years ago & homosexuality is not as a problem as it was back then, health information is still very confidential and not as secure as you might think. During May 2017, the attack of the malware “Wannacry” infested computers of the British national health service, making data inaccessible. With patientory, this incident would maybe not had endangered the data of patients.
However: from my experience, healthcare companies are quite difficult to work with. The mentality of “never change a running system” is – at least in Germany – so heavily entrenched in the minds of the executives, that it is hard to get healthcare related companies to adapt to the digital age. Let alone to high-tech like blockchain tech.
Well, let’s see if PTOY can deliver on their promise.
Patientory is a registered company in the US, with Chrissa McFarlane as the CEO. She founded the company back in 2015, with it being outside of public awareness until 2017.
Their website itself looks professional & offers a huge amount of information, including:
- team composition
- focus group centered “what our technology do”
- media coverage
- blog
- quite big FAQ section
And other subsections, that makes initial research on Patientory quite an easy feat.
For the team, it is necessary to go back into the past a bit. Because I remember a huge load of criticism being thrown at McFarlane, when PTOY entered the market back in June. Reason was not only because of her being a woman in the male dominated industry of blockchain projects, but because of her qualifications – she has a bachelor of arts in African Studies & Spanish. Objectively, this is not something one would automatically deem to be qualifying to be a CEO of a high-tech startup. However: as in many cases of FUD, this is only half the truth, as she also has a Master degree in Business, focused on “Marketing Research & Healthcare”. A degree in another field or even not having a degree at all is not necessarily a KO-criteria per se, as some successful CEOs also did/do not have a degree (one of them being Steve Jobs). In the case of McFarlane, it is indeed not as bad as it is advertised by many critics, as she has a Master in business. Still – it is rare to see someone with a business degree working on this, as most other CEOs or project leaders I encountered are from a technological background.
The team is listed with their LinkedIN accounts, that are all quite detailed. Their qualifications are spread out over several fields – software development, marketing, consulting, medical expertise etc. I however fear the team itself might be too small and not focused enough on the project. Because from the 6 persons listed there, the qualifications are spread like this:
- two with experience in IT development (Michael Beer & Rob Campbell)
- one for UI/front end development (A’nita Evans)
- one for project management (Jennifer Georgino)
- one for marketing (Palakh R Saraogi)
- one CEO (Chrissa McFarlane)
Please do not misunderstand me: project management, marketing and strategic decision making is heavily important for a company – missing out on these can be crippling. Take KMD as a case for bad marketing (messed up presentation) and how it impacted the price of their currency. But having only 2 developers with actual competences to actively develop the technological foundation out of 6 employees in total is not exactly, what I would expect from a high-tech startup. Especially, when they raised 7.2 million USD during their ICO, which should make them able to hire more than enough programmers.
However: as their website is advertising other positions, one of them being an offer for a full-time position as a software developer, it is not to be expect the team to stay this small. As the people working on Patientory 1st seem to be experienced in organizing projects and are 2nd open about their plans (I talked to them on slack about the team composition & my fears of it), I would not consider this a complete KO-criteria. It however still worries me a bit, as I do not see, how they use these funds right now.
Patientory received good publicity by the blockchain news outlets, as well as some tech focused media outside of blockchain technology. Not everyone of them is purely praising them however, some of them are straight away pointing out legal issues. It shows that they seem to be aware of possible problems, they may encounter in the future (actual legal issues are the problem, that came to my mind, when I first did research on Patientory).
They list 6 partners on their website:
startup +health
This partner seems to be a company incubator/accelerator, focusing on healthcare related IT startups to modernize healthcare and help it to adapt to digitalization, as well as interconnecting the companies with each other and global players in this field. It is great to see a partner like that among them, as it perfectly matches their goal – but I would not put an emphasis on this partner. While it is good to be in a big network, it is not solely dedicated to Patientory after all.boomtown health tech accelerator
An accelerator which program Patientory seems to have passed through.life science Washington
Another accelerator.Atlanta Tech Village
The accelerator whose program Patientory is partaking right now.Womens startup lab
Another accelerator, but Patientory did not seem to partake in the accelerator phase.Atdc Georgia Tech
And another incubator, whose services Patientory still seem to use, as they are not listed among the Alumni.
Here, I am a bit in between two positions regarding how to look at these partners. On the one hand, it is quite normal for a startup to pass through several incubator/accelerator programs. It is definitely not a bad thing, quite the opposite. These programs are great for building up a network, get access to funds, low interest loans, cheap premises, gaining know-how about corporate governance and many more things, that are very beneficial for any young company. Such programs strengthen a young company in its most vulnerable state: while it is being funded and doing first steps into the real world.
But I need to point out, that these partnerships are not specific enough for Patientory to have a significant impact, to attract new investors. A partnership with a hospital for a pilot project would be major. A partnership with an EMR-company (electronical medical record) would be major. But incubator/accelerator programs “only” create a solid foundation for the company to build on. Very good, but too general to be significant.
One partner not listed on their website came up just recently: the blockchain project DASH. Patientory entered their program to make both blockchains interoperable, making it more usable for future users of DASH to use the services of Patientory. This is indeed a major partner, but unfortunately does not free them from the need of partners in the healthcare industry. Patientory is not a blockchain industry focused project like Komodo or ARK and any sustainable future value of Patientory is based on how, when and to which extend they are able to enter the world outside of blockchain industry.
Publicity: 6.5/10 (6 – good & solid fundamentals, but Patientory needs more actual partners inside the health industry aside from the accelerator/incubator programs they passed through/partake right now; +0.5 for the excellent press coverage outside and inside of the blockchain community; +0.5 for the recent partnership with DASH; -0.5 for the lack of transparency in regard what they do with the money, as it does not seem to be used on additional staff)
They have a twitter account, one subreddit, slack, as well as a telegram group, a blog on their own website as well as a YouTube channel. The slack is active and quite usable, but their telegram link seems to be dead, meaning either the Telegram chat is not in use anymore or the link is dead (will not influence the grading, as Telegram is not a common communication platform for blockchain projects). The subreddit is not as active as the one of ARK, but after all – Patientory does not follow the same democratic philosophy of ARK and therefore does not quite require the same exceptional degree of community interaction. They conducted two AMAs so far, which can be viewed in full length on their YouTube channel. Further AMAs in the future would be a good idea, as it keeps the community up-to-date on the project as well is appealing to new investors.
Patientory also has a constantly updated roadmap and a whitepaper. The whitepaper is a bit abstract and focuses on a critical analysis on how blockchain technology may be used to create a HIPAA-act (Health Insurance Portability and Accountability Act) compliant product, a schematic outline in how this will be achieved on a technical level, how PTOY as a token will be used in the future product, as well as how it will benefit the patient, their health care providers and the doctors in the future.
The roadmap is a mixture of future targets as well as past accomplishments and milestones. I am not so much of a fan to put past media coverage and headlines like “ranked top 5 in…” in the roadmap. Also for those, who are interested in the actual project progress, it makes the roadmap harder to read. However, it is not unreadable and so far, every past cornerstone seems to have been reached. One point of minor criticism is, that these steps are not pinpointable on specific dates, only on months – in general, this is fine. But for major points (like the release of their app), I personally would prefer more exact dates.
Other than that, I expect them to update their roadmap in the near future (around October), as they run out of upcoming milestones on it and it would leave them with only an outdated roadmap with vague goals for the whole year of 2018.
Right now, Patientory is traded at Bittrex, Liqui, Etherdelta and HitBTC. Bittrex takes the majority of about 90% of the whole trading volume, ~9% is found on Liqui and only minor amounts on Etherdelta and HitBTC. So, the price is yet again mostly dependent on the trading on Bittrex.
Activity - grade: 6.5/10 (6 – active communication channels that could see a bit more community involvement; +0.5 for the solid whitepaper that also addresses specific legal issues)
Let’s get to the problem, Patientory wants to solve in detail. I am sorry that it might be a bit longer – but because most people I talked with about Patientory, did not seem to grasp what they intended to do.
Like I wrote in the introduction already, release of medical data of an individual can be devastating for this person. Therefore, these information are highly confidential and only to be handed out to the responsible caretaker, if it is necessary for them to look at the medical records of their patient. These persons having these information is likely to be beneficial for the convalescence of a patient or can at least save the time and money for everyone involved. This way, everyone can focus on their role: the physician to take care of the patient, the patient to get up on his feet again and the hospital to organize this whole procedure.
Now, we run into a conflict: the data is highly confidential but needs to be available to many people, to put their knowledge to the most effective use. If one person needs a nutritionist and a rehab nurse to adjust his diet after an operation as well to get rehab, we have a huge amount of people and organizations having to deal with this data:
- the hospital with the surgeon, nurses, administrative staff etc.
- the nutritionist
- the primary physician of the patient
- the health insurance provider
- the labs providing further data
- the pharmacist
- the rehab nurse
Most of this data is stored in centralized (siloed) storages e.g. the hospital itself. But the nutritionist might need full detail on the medical history of his patient, needing to not only consult the hospital because of the operation but also the primary physician of the patient. Same for the rehab nurse. And the health insurance provider most likely needs detailed information about anything from anyone involved. In the end, someone needs to send this data to them or establish a secure connection in between, to get this done.
But as these storages are centralized, the exchange between these storages not only take a while, but opens them up for a significant danger for hackers & malware, who are aware that health data is highly valuable. On top of that, the caretakers need to be involved with this strictly administrative procedure, meaning that they have an additional task to do, besides focusing on their job, to get their patient back to good health. In the end, we are left with higher administrative expense, badly allocated work force and overall a healthcare, that is not solely focused on getting a human being back to good health. A physician should take more care of his patients than he does administrative work.
We also run in another problem with this whole system of centralized data storages: the patient himself is giving away the control of his data. The more his data is accessed by parties outside of his control, the higher the danger of a breach of personal information.
Now it is easy to summarize, what they aim to solve:
For the patients: | For the service providers: | For science: |
---|---|---|
Securing confidential health information on the blockchain and putting it back into his hands | Secure storage of the medical history of a patient as well as access to it | Easy access to depersonalized health data for research purposes |
Enable him to communicate directly with the persons | Lower expenditure to secure patient medial data (Patientory promises ~42% decrease – the German controller inside me would really like to know, how this number was calculated) | --- |
However, we run into a bit of a problem here, when we want to analyze details in the technology. As the service of Patientory falls under the HIPAA (Health Insurance Portability and Accountability Act), the service needs to match certain regulations to be able to be used in the first place. It is impossible to circumvent this, as the Hospitals and caretakers need to use compliant tools, to store and handle patient information.
This also is among the reasons, why Patientory does not disclose any source code. Their Github is empty, except for their whitepaper. The other reason being security, which leaves me a bit puzzled.
Because after consulting some coders among my wider circle of friends, most of them told me, that there is no direct connection between security and open-source. I am not versed in American regulation in regards of compliance to HIPAA, so I do not know, if the most important reason for not disclosing source code is compliance or not.
However, there are also several open source EMR systems like Care2x, OpenEMR and OpenVista, which are open source – so the argument regarding security and regulations does seem a bit strange for me. But like I said: I am no professional in US health care regulations. This very well might be, because they want to avoid delaying their market entry. In this case, this decision is very good, as it fastens them getting traction. But why is it not openly communicated this way?
Another issue that sprang into my eye, was this passage on how Patientory will be beneficial for service providers:
I admit, I am – at best – a hobby-coder. For me, programming an AI is most likely as impossible as understanding advanced theoretical physics. And I could not find anyone on their team or among their advisors, who has had anything even remotely to do with AI programming in his/her past. Their services have nothing to do with what the AI should be able to do. I would not consider this to be a major issue, if the task issued on the “AI” would not be such a crucial one. But interconnectivity in between completely different systems is a) not actually a trivial task and b) indeed crucial to make Patientory appealing for hospitals to use it in the first place. In German hospitals, you sometimes encounter technology that is quite old and/or an isolated application, exclusively developed and supported for a handful of hospitals – or even worse, a single one.
My point here is, that a major issue in the appeal of Patientory – fast transition of data from the expensive legacy EHR database to Patientory – may be solvable via usage of an AI. But AI is not something, that you see in the team or the whitepaper right now. There are three sections, that list something about AI: the “how it works” section, one point of the job advert for a software engineer and in the roadmap (“machine learning”). I know that Patientory aims also to use AI for shortening the time for diagnosis, but I start to get a feeling that Patientory wants to solve too much, without having anything to show off.
It is not that I doubt their dedication to the project. Or that the tech is not able to achieve their goal in general. It is just, that I get the feeling, that they either a) fail to communicate enough to ensure a good level trust of the investors or b) underestimate the efforts needed, to accomplish their goals. If I (as someone who digs only a bit deeper in their business model for just about a week) has doubts in these fundamental questions, how do they want to persuade the executive of a big health company to use their technology?
Technology - Grade: 4/10 (4 – the technology is not gradable, as there is basically no tech public, making it hard to believe in the tech in the first place)
Patientory is traded for about 4 months now and encountered a major pump, right after completing their ICO and the beginning of trading on Bittrex. During this increase, the price reached its ATH and entered a bearish phase right afterwards, which culminated in an all-time low of about ~4,200 satoshi.
The price since then had slowly and steadily increased to levels of about 8,000 satoshi per coin. That is until the September 11th, when the price almost doubled within 2 days. This was most likely due to the hype around the participation of Patientory at the Venture Atlanta 2017 – a venture capital networking event in the US, where Patientory was selected as one of the 35 presenters. Since then, the price seems to have stabilized at around 10,000 satoshi.
The team itself was not aligned with any other former blockchain project. However, some of the former members left the team. Michael E. Rubin (who is still listed as the contact for press inquiries on their website) left the team to work for another company (Vezt inc.). He was the one, who moderated the AMAs back in June and July.
Another person having left the team of Patientory, to work at Vezt inc. recently, is Jesse Brown. This is worth mentioning, as he had the quite important position in Patientory: he was the CTO (Chief Technology Officer). He seems to be well versed in the development of blockchain technology – as long as Patientory does not name a replacement as a CTO for him, I would consider this an issue. If not for the development, then for the trust of the investors in the project. He seems to be replaced by Rob Campbell, who have joined Patientory after the ICO. He however does not list such an extensive experience with blockchain technology like his predecessor, according to his LinkedIn account. Also, if he indeed took the position of the CTO, it would be very good for them to communicate this – because right now, he is the “Chief Software Architect” or “Chief Solutions Architect”, depending on where you look it up.
I could not find major delays in their roadmap so far. What worries me however, is a small detail that changed in between the current version of the roadmap and the one before. In the older one, the app was to be released in September 2017. In the updated roadmap from the 20th of September, the app will be given to the FDA (Food & Drug Administration, a US agency handling a wide range of regulations regarding… well, food and drugs) to get approval for the release. While it means that the app seems to be done, it also hints to the fact, that the need of FDA approval was not taken into account before. In any case: the app will be most likely not released in September, meaning they will fail to deliver on their roadmap – which is not so bad, if it would be communicated properly.
Another major thing I found was quite an unprofessional answer to someone on reddit. What I mean is the thread “7 red flags”, to whom Patientory answered not on reddit, but on their own website.
Problem in this case is not the initial post. Because even if we consider these “7 red flags” as simple FUD, which only aims to harm the reputation of Patientory (which is not the case, as it points out some issues you also can read in this analysis), the answer on their website was simply not professional. If you think it is FUD, you either answer it polite and based on arguments – or you simply ignore it. I guess (hope?) they learned from it, as the answer to “7 red flags” is not available on their website anymore. Such happenings severely hurt the trust of the investors, as it hints to them not being able to handle criticism well.
History - grade: 4.5/10 (5 – roadmap is there, but the most recent change of the next upcoming major milestone (app) was suddenly changed with no sufficient explanation – if this happens more often, a grading of 4 is inevitable; -0.5 for the unprofessional response during the ICO)
Next upcoming major thing would have been the app, which was announced to be released in September. The app seems to be done already, but is now awaiting approval of the FDA. It is not possible to predict exactly, when the FDA will be done reviewing it – and in the roadmap of 2017, no other point of comparable impact can be found. After some research, I found a list of other applicants for approval of their mobile medical applications (MMA) on the website of the FDA itself. The average time for them to process an application from submission until approval is in between 2-4 months for most other projects I checked. This would mean, that the actual release of the app is expected to be pushed back to 2018 in the worst case.
Again: I am no professional in regards of US health regulations. I might have found the wrong information in this regard & Patientory might fall into another regulation, which can be processed way faster. If so, please contact me on twitter.
Right now, 100,000,000 PTOY are in existence (no further coins will come in existence):
- 70,000,000 sold during their ICO (in circulation)
- 20,000,000 allocated in the Patientory Foundation for funding research & development (not in circulation)
- 10,000,000 go to the founders to be given out when meeting milestones over the next 5 years (not in circulation)
So, it is not very likely, that a major amount of PTOY will suddenly enter the market, crashing the price. Also, I like their incentive system for their employees to pin the incentive payment of PTOY to milestones – if it would be a bit more transparent. Because it would be nice to know, which milestones are needed to be reached at which point in time.
Right now, Patientory does have some competitors in regard of healthcare & blockchain technology, namely:
- PokitDok (who seemingly want to do the same as Patientory, but is a company that has no currency right now – they still aim to use Blockchain technology and have Intel as their partners, but seem to not utilize an ICO but normal investors to fund themselves)
- UmbrellaCoin (who focuses more on providing an insurance system based on smart contracts, which Patientory also aims to be able to do later on)
- MedCredits (who want to focus on telemedicine first & aim to expand into being an EMR – their ICO ended about 10 days ago, but they are not listed anywhere for now)
It is hard to tell, how Patientory stands in between these competitors, as most of them are either too young (MedCredits/Umbrellacoin) or too intransparent (PokitDok) to be properly compared with Patientory. So far, Patientory is therefore the biggest blockchain project with a currency freely traded on exchanges.
The future development in respect of the targeted industry is, where some people tend to overestimate the market. The overall coverage of EMR systems among all health care providers is about 67% (growing only ~1% from last year). Office-based physicians on the other hand completely adopted EMRs already, with a quote of about 90%. This means, that they are entering an industry that might be already too saturated with EMR-services. The growth might be too small for them to compete effectively with already established EMR providers.
Future - Note: 5/10 (5 – the future does have some obstacles, which will be hard to overcome; they however right now have a head start as the first blockchain based healthcare project, which they need to use – otherwise, I doubt that they will have the impact they want; there is no immediate news that will influence the price significantly in the foreseeable future)
- Publicity: 6.5/10
(6 – good & solid fundamentals, but Patientory needs more actual partners inside the health industry aside from the accelerator/incubator programs they passed through/partake right now; +0.5 for the excellent press coverage outside and inside of the blockchain community; +0.5 for the recent partnership with DASH; -0.5 for the lack of transparency in regard what they do with the money, as it does not seem to be used on additional staff) - Activity - grade: 6.5/10
(6 – active communication channels that could see a bit more community involvement; +0.5 for the solid whitepaper that also addresses specific legal issues) - Technology - Grade: 4/10
(4 – the technology is not gradable, as there is basically no tech public, making it hard to believe in the tech in the first place) - History - grade: 4.5/10
(5 – roadmap is there, but the most recent change of the next upcoming major milestone (app) was suddenly changed with no sufficient explanation – if this happens more often, a grading of 4 is inevitable; -0.5 for the unprofessional response during the ICO) - Future - grade: 5/10
(5 – the future does have some obstacles, which will be hard to overcome; they however right now have a head start as the first blockchain based healthcare project, which they need to use – otherwise, I doubt that they will have the impact they want; there is no immediate news that will influence the price significantly in the foreseeable future)
Overall score: 5.3 / 10
Patientory is a project that leaves me in between two opposing positions.
On the one hand, Patientory is a young inspiring start-up, that wants to utilize the blockchain to create a more efficient & more safe health care system. This is beneficial for everyone involved: patients, insurance providers, scientists, physicians, hospitals etc. A very noble goal indeed, which makes it very easy, to fall in love with the project. I for sure did, while I did research on it. And I would love to see them prosper.
On the other hand, there are points they need to adjust. First, there is the need of more intense communication with potential investors, who are interested in PTOY, but have not bought it yet. Then, a not well communicated substantial change of plans on their roadmap raises questions (“release of the app” transformed to “app submission to FDA” without proper announcement). And on top, I have the general impression, that they underestimate what a dedicated community can do, to help them achieving their goal.
Because the people buying crypto currencies are from every background thinkable – lawyers, construction workers, nurses, doctors, housewives, students, professors, programmers etc. Not everyone of them just want to buy low & sell high, without knowing what the project does or caring what it will do, after they sold their coins. Some get attached to the project and want to help – regardless with what little it may be, what they can contribute. This is something they need to address, otherwise I see their chance to success diminished by a substantial degree. At least a detailed public communication, why something cannot be disclosed would be great.
Regarding of price: right now, PTOY is finding its support, after the hype of their presentation at Atlanta Venture. I see a big probability them dropping back to ~8,000 satoshi per coin, as the supposed pushback of the release of the app in the end leaves no major event in the foreseeable future, to justify a higher price. And with no major happening, I expect it to stay on this level and only slowly getting traction, if they do not change their course of action.
Which in the end might be too slow, if their competitors show up on the market.
PDF with footnotes and sources
My blog
Originally released on twitter one week ago
If you like my analysis & want to give something back – vote here on steemit or maybe treat me a beer with BTC.
After all, I am a German penguin.
BTC: 12dTyxchdGhYjGBi1QFVPWbagQrRSJssWT
ETH: 0x3D8e6B27F7ab389888791ABEe6FA62F4718A1164
LTC: LT6qfsrxPhVxcYPbNgT267W1aVRd6n5AFq
One I've invested in. Would love to see it succeed for reasons other than my wallet. 😎