Tag Archives: statistics

Research Reporting Bias in Clinical Trials

As our healthcare system shifts to corporate ownership, the drive to generate profits is destroying healthcare (and all other social services).

When decisions are made on a financial basis instead of patient welfare, inappropriate standardization is applied where human variety is critically important: the fluctuating biochemistry of our individual bodies and how they react to interventions.

Here are 3 PubMed studies showing how research is corrupted by financial motives:

Frequency and reasons for outcome reporting bias in clinical trials: interviews with trialists – free full-text /PMC3016816/ – Jan 2011   Continue reading

CA Links Past Opioid Rx by Doctors to Deaths

California Doctors Alarmed As State Links Their Opioid Prescriptions to Deaths – Heard on All Things Considered – Jan 2019

Of all the hare-brained ideas states have implemented to clamp down on legal and medically appropriate opioids, California has come up with the worst:

Comb through the state’s PDMP for the last three years to find a link from someone who recently died from an overdose of illegal drugs to a doctor who prescribed them opioids up to three years ago, and threaten them with prosecution if the CA medical board doesn’t agree with some aspect of this long-ago prescription.

About a year ago, Dr. Ako Jacintho of San Francisco returned home from traveling to find a letter from the state medical board waiting for him.   Continue reading

The Trouble with Meta-Analyses

The trouble with Meta-Analyses – Thoughts About How They Should Contribute to Medical Science and Practice – by Milton Packer – Nov 2017

Many have critically examined the methodology of meta-analysis, and others have set standards for their execution. Despite such guidance, meta-analyses continue to proliferate, but we should ask: do they really contribute?

Esteemed organizations regard the conclusions of a well-executed meta-analysis as a higher level of evidence than a single well-done clinical trial.

This commentary explains why this cannot possibly be true.   Continue reading

Curbing Rx opioids will have little effect on overdoses

Curbing Rx opioids will have little effect on overdoses – comments from Stefan Kertesz on Twitter @StefanKertesz – Feb 2019

I previously posted about the study that just came out which estimates the future trajectory of the “opioid crisis” and finds little use for restricting legitimate prescriptions:  Study: Limiting opioid Rx won’t reduce overdose deaths

1/With 3% of overdose deaths (9 of 269) in Jefferson Cty, Alabama due to prescription opioids, there is a limit to how many lives are going to be saved by extending the 7 year decline in Rx, as underscored in a new study by ⁦@LarochelleMarc⁩ https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2723405

Continue reading

Over-Restricting Pills = More Deaths

Dear PROP/CDC, Here’s What Happens When You Over-Restrict Pills: More Deaths. Nice Going. – By Josh Bloom — Dec 2018

If Shakespeare were alive today he would be hard-pressed to come up with a script that could match the tragedy that has been imposed on this country by

  • self-appointed drug experts,
  • bureaucrats,
  • self-serving politicians, and
  • various other fools.

It’s that bad. And it was largely preventable.   Continue reading

Almost All Overdose Deaths Involve Multiple Drugs

Almost All Overdose Deaths Involve Multiple Drugs, Federal Report Shows  – By Erin Schumaker – Dec 12, 2018

A new report analyzing the drugs involved in fatal overdoses once again emphasized fentanyl’s role in the United States’ opioid crisis and highlighted a point frequently made by public health experts:

Most people who fatally overdose
have more than one drug in their system.

In 2016, about 70 percent of fatal overdoses involving fentanyl or heroin involved another drug as well, and roughly 74 percent of fatal overdoses involving cocaine also involved one or more other drugs.   Continue reading

Today’s Drug Abusers Not From Yesterday’s Patients

Today’s Drug Abusers Did Not Derive From Yesterday’s Patients – By Jeffrey A. Singer – December 4, 2018

We learned last week that the 2017 drug overdose numbers reported by the US Centers for Disease Control and Prevention clearly show most opioid-related deaths are due to illicit fentanyl and heroin, while deaths due to prescription opioids have stabilized, continuing a steady trend for the past several years.

But the media and policymakers remain unshakably committed to the idea that the overdose crisis is the product of greedy pharmaceutical companies manipulating gullible and poorly-trained doctors into over-prescribing opioids for patients in pain and ensnaring them in the nightmare of addiction.   Continue reading

Chronic Pain Patients Did Not Cause Opioid Epidemic 

Chronic Pain Patients Did Not Cause Opioid Epidemic – by Roger Chriss – May 2017 (Re-post)

By carefully cross-checking the numbers, Mr. Chriss proves it impossible to find patients responsible for the “illicit opioid crisis”.

Contrary to common belief, chronic pain patients are not all opioid addicts and did not cause the opioid crisis. The vast majority of patients who are prescribed opioids rarely misuse or abuse them.

Opioid addiction is real and should not be ignored or downplayed, but we need to identify its true causes. Despite the growing number of restrictions on prescription opioids, overdoses and related deaths continue to rise, which strongly indicates that pain patients have very little to do with the so-called epidemic.  Continue reading

How to lie, cheat, manipulate, and mislead with statistics

How to lie, cheat, manipulate, and mislead using statistics and graphical displays – by Rick Ord, rord@ucsd.edu – 2010

This is a humerous slide presentation (with Dilbert!) about the malfeasance we must endure in so much of “scientific” research lately, especially when the results must be forced to support the predetermined “idea” about the study (as explained in my recent post: Statisticians Asked to Commit Scientific Fraud).

Here’s just a quick list of some of the shenanigans he covers:

  • Mislead with poor analysis,
  • Mislead with averages,
  • Mislead with poor graphical displays
  • Mislead using chart adjustments
  • Mislead using bar charts

I highly recommend you give it a look: How to lie, cheat, manipulate, and mislead using statistics and graphical displays


Researcher Requests for Inappropriate Analysis and Reporting

Researcher Requests for Inappropriate Analysis and Reporting | Annals of Internal Medicine | American College of Physicians | Oct 2018

Here is the verbatim abstract of a sad truth about the corruption of scientific research by, what else, money.

The previous post, Statisticians Asked to Commit Scientific Fraud, discussed the results in more detail and I’m showing this is mainly as a reference. If anyone has access to the full paper, I’d love to see it.

Inappropriate analysis and reporting of biomedical research remain a problem despite advances in statistical methods and efforts to educate researchers. Continue reading