Connecting the dots

No man is an island, entire of itself…. Each man’s death diminishes me, for I am involved in mankind (John Donne, 1624)

Connected
Connected

Einstein was of the same mind in writing: “A human being is a part of the whole, called by us ‘Universe’, a part limited in time and space. He experiences himself, his thoughts and feelings as something separated from the rest — a kind of optical delusion of his consciousness.” At this point opinions diverge as to whether his letter goes onto search for meaning in religion, or release from confinement in a delusional prison through compassion. Nonetheless, from poets to scientists, it’s obvious that we all have a need to be connected. That’s in sharing communities, and not just online 24/7 in Zuckerberg’s empire 😉     The eternal battle to subdue isolated individuality is most apparent when we struggle with illness – Maslow put physiological and safety needs as a priority ahead of belongingness, and indeed health and wellbeing take our focus when in pain.

In the last few years Patient Reported Outcomes (PROs) collect information on the impact of illness, and gather data on solutions. Every ‘breakthrough’ story on social media will draw a following and perhaps develop a market, but it’s only when all success stories are aggregated that we can discern gold-raking from true nuggets. An example allows your experimentation with this idea by querying the database for other’s experiences, leveraging nearly half a million subscribers.

My other blog demonstrates a practical use for the power of community. Their premise at PatientsLikeMe is to donate your data to find a cure, realistically however this is underhand collection of feedback on medications by their sponsors. Nonetheless, if you subscribe and use their search facility to connect with others who use meds that may have been recommended to you, or have similar bloods pathology, or experience like impact to QualityOfLife…… then you allow the likes of me to extract rankings of treatments by efficacy. This appears to be an unintended consequence, since when I thanked the administrators for their data one query window was promptly closed. But that type of portal opening provides weak observations only.

There is no way to control for the confusing of data due to an individual’s choices. If, as with Carole & I, the PRO subscriber is interested in mind and body, then other Complementary and Alternative Medicine therapies are likely to be used also. Who can say which of these multi-disciplinary treatments is the most beneficial? The solution to this is medicine’s gold standard for evidence, the Randomised Clinical Trial. Participants are randomly allocated regardless of their disposition to receive either the intervention under investigation, or else standard care. They, and the trial staff, are blinded as to which group the coin-toss has placed them in, so as to preclude bias. This is vital when the measured outcome is subjective, as described by our Indian friends.

A significant difference between group’s results means that it was a strong enough result that it was improbably due to chance (eg four heads in a row is surprising when head or tail results are equally likely, but possible with odds of 1 in 16). A cutoff point of 1 in 20 is universal, written as p=0.05 or less. To get a significant result when the difference between control and intervention is small requires larger studies (more tosses of the coin), since halving precision from say p=0.1 means quadrupling the participant count. An inadequately budgeted trial will often be a pilot study, so as to prove the worth of that fourfold funding increase.

The last public grants round from NHMRC only approved one sixth of applications. Industry often makes up for this shortfall, and minor benefits can be shown as significant with their larger budget 😒     Blinding now allows corruption of results, as secrecy is applied to the identities of peers who’re reviewing articles for probity and accuracy. That link explodes some fallacies associated with concealment, a good suggestion by biostatisticians  but one that’s subject to cunning workaround.

Blind Lady Justice - consigned to mythology?
Blind Lady Justice in mythology – consigned to oblivion ?

Only universities are allowed to conduct such research, and it’s highly regulated. Accelerated understanding of biomedicine from the laboratory (in vitro) or in animals (in vivo) has set a pace that human drug trials have been unable to match. Breakthroughs from reckless self-administration early in the 20thC, then followed by cruel and callous human experimentation in the middle years (including the Tuskagee trials on Afro-Americans along with the holocaust atrocities), means bureaucratic oversight of research has grown to verge on a totalitarian structure. There’s many complaints of staleness, of copycat ‘me-too’ or copyright extending ‘me-again’ drugs. All done behind closed doors.

It is to be hoped that the online trend to open source communities – some examples being Android/Linux computing, Creative Commons licensing, or Public Library of Science publishing will encourage further usage of other PROs than PatientsLikeMe. Outside the system, and managed by those whose self-interest is patient welfare rather than career.

 

 

 

Advertisements