One Example of Healthcare Misinformation

On June 21^st  NBC News aired an investigation into HCA Healthcare accusing HCA administration of pressuring doctors, nurses and family to have patients enter hospice care or be discharged (1). Patients entering hospice care can lower inpatient mortality rate and length of stay, increasing profits and bonuses for executives. It works this way — if a patient passes away in a hospital, that death adds to the facility’s inpatient mortality figures. But if that person dies after a transfer to hospice care — even if the patient stays at the same hospital in the same bed — the death doesn’t count toward the facility’s inpatient mortality rate because the patient was technically discharged from the hospital. A reduction in lengthy patient stays is a secondary benefit according to an internal HCA hospital document (1). Under end-of-life care, patients don’t typically live long, so the practice can allow HCA to replace patients that may be costing the facility money because their insurance has run out with those who generate fresh revenues.

These practices are not unique to HCA nor are they new. Manipulation of patient data such as mortality go back at least until the 1990’s. For example, at the Phoenix VA the floor inpatient mortality rate was low while the ICU mortality rate was high. This was apparently due to excess mortality in floor to ICU transfers (2). Reduction of inappropriate ICU transfers from the hospital floor corrected the high ICU mortality rate. Similar changes were seen for length of stay. There were also dramatic reductions in the incidence of ICU ventilator-associated pneumonias and central line-associated blood stream infections just by alternating the reported cause of pneumonia or sepsis. For example, ventilator-associated pneumonia was called “delayed onset community acquired pneumonia” and sepsis was blamed on a source other than the presence of a central line.

These data manipulations were not restricted to the inpatient mortality or length of stay. Outrageously exaggerated claims of improvement and lives saved became almost the norm. In 2003 Jonathan B. Perlin, then VA Undersecretary of Health, realized that outcome data was needed for interventions such as pneumococcal vaccination with the 23-polyvalent pneumococcal vaccine. On August 11, 2003 at the First Annual VA Preventive Medicine Training Conference in Albuquerque, NM, Perlin claimed that the increase in pneumococcal vaccination saved 3914 lives between 1996 and 1998 (3) (For a copy of the slides used by Perlin click here). Furthermore, Perlin claimed pneumococcal vaccination resulted in 8000 fewer admissions and 9500 fewer days of bed care between 1999 and 2001. However, these data were not measured but based on extrapolation from a single, non-randomized, observational study (4). Most studies have suggested that the 23-polyvalent vaccine is of little or no value in adults (5).

It raises the question of why bother to manipulate these data? The common denominator is money. Administrators demand that the numbers meet the requirements to receive their bonuses (1). At the VA the focus changed from meeting the needs of the patient to meeting the performance measures. HCA administration is accused of similar manipulations. Speculation is that many if not most healthcare administrators behave similarly. The rationale is that the performance measures represent good care which is not necessarily true (5).

Who can prevent this pressuring of care givers and patient families to make the numbers look better? One would expect that regulatory organizations such as the Joint Commission, Institute of Medicine, Centers for Medicare and Medicaid Services, Department of Health and Human Services, and Department of Veterans Affairs would require the data reported be accurate. However, to date they have shown little interest in questioning data which makes their administration look good. The Joint Commission is a National Regulatory group that is prominent in healthcare regulation. After leaving the VA in 2006, Perlin was named the President, Clinical Operations and Chief Medical Officer of Nashville, Tennessee-based HCA Healthcare prior to being named the President and subsequently CEO of the Joint Commission in 2022. When regulatory organizations get caught burying their heads in the sand, administrators usually respond by blaming the malfeasance on a few bad apples. An example is the VA wait scandal that led to the ouster of the Secretary of Veterans Affairs, Eric Shinseki, and the termination of multiple administrators at the Phoenix VA. It should be noted that although Phoenix was the focus of the VA Inspector General at least 70% of medical centers were misreporting the wait times similarly to Phoenix (6).

Who should be the watchdogs and whistleblowers on these and other questionable practices – obviously, the hospital doctors and nurses. However, the hospitals have these employees so under their thumb that any complaint is often met with the harshest and most severe sanctions. Doctors or nurses who complain are often labeled “disruptive” or are accused of being substandard. The latter can be accomplished by a sham review of patient care and reporting to the physician or nurse to a regulatory authority such as the National Practitioner’s Databank or state boards of medicine or nursing (7). Financial data may be even easier to manipulate (8). A recent example comes from Kern County Hospital in Bakersville, CA (9). There the hospital’s employee union accuses the hospital of $23 million in overpayment to the hospital executives over 4 years. According to the union the hospital tried to cover up the overpayment. Now the executives have requested the hospital board to cover the overpayments.

The point is that hospital data can be manipulated. One should always look at self-reported data with healthy skepticism, especially if administrative bonuses are dependent on the data. Some regulatory authority needs to examine and certify that the reported data is correct. It seems unlikely that Dr. Perlin’s Joint Commission will carefully examine and report accurate hospital data. Hopefully, another regulator will accept the charge of ensuring that hospital data is accurate and reliable.

Richard A. Robbins, MD

Editor, SWJPCCS

References

1. NBC News. HCA Hospitals Urge Staff to Move Patients to Hospice to Improve Mortality Stats Doctors and Nurses Say. June 21, 2023. Available at: https://www.nbcnews.com/nightly-news/video/hca-hospitals-urge-staff-to-move-patients-to-hospice-to-improve-mortality-stats-doctors-and-nurses-say-183585349871 (accessed 6/28/23).
2. Robbins RA. Unpublished observations.
3. Perlin JB. Prevention in the 21^st Century: Using Advanced Technology and Care Models to Move from the Hospital and Clinic to the Community and Caring. Building the Prevention Workforce: August 11, 2003. First Annual VA Preventive Medicine Training Conference. Albuquerque, NM.   
4. Nichol KL, Baken L, Wuorenma J, Nelson A. The health and economic benefits associated with pneumococcal vaccination of elderly persons with chronic lung disease. Arch Intern Med. 1999;159(20):2437-42. [CrossRef] [PubMed]
5. Robbins RA. The unfulfilled promise of the quality movement. Southwest J Pulm Crit Care. 2014;8(1):50-63. [CrossRef]
6. Department of Veterans Affairs Office of Inspector General. Concerns with Consistency and Transparency in the Calculation and Disclosure of Patient Wait Time Data. April 7, 2022. Available at: https://www.va.gov/oig/pubs/VAOIG-21-02761-125.pdf (accessed 6/28/23).
7. Chalifoux R Jr. So, what is a sham peer review? MedGenMed. 2005 Nov 15;7(4):47; discussion 48. [PubMed].
8. Beattie A. Common Clues of Financial Statement Manipulation. Investopedia. April 29, 2022. Available at: https://www.investopedia.com/articles/07/statementmanipulation.asp (accessed 7/28/23).
9. Kayser A. California Hospital Accused of Overpaying for Executive Services. Becker’s Hospital Review. June 28, 2023. Available at: https://www.beckershospitalreview.com/compensation-issues/california-hospital-accused-of-overpaying-for-executive-services.html?origin=BHRE&utm_source=BHRE&utm_medium=email&utm_content=newsletter&oly_enc_id=6133H6750001J5K  (accessed 6/29/23).

Cite as: Robbins RA. One Example of Healthcare Misinformation. Southwest J Pulm Crit Care Sleep. 2023;27(1):8-10. doi: https://doi.org/10.13175/swjpccs029-23 PDF

The Need for Improved ICU Severity Scoring

How do we know we’re doing a good job taking care of critically ill patients? This question is at the heart of the paper recently published in this journal by Raschke and colleagues (1). Currently, one key method we use to assess the quality of patient care is to calculate the ratio of observed to predicted hospital mortality, or the standardized mortality ratio (SMR). Predicted hospital mortality is estimated with prognostic indices that use patient data to approximate their severity of illness (2). Examples of these indices include the Acute Physiology and Chronic Health Evaluation (APACHE) score, the Simplified Acute Physiology Score (SAPS), the Mortality Prediction Model (MPM), the Multiple Organ Dysfunction Score (MODS), and the Sequential Organ Failure Assessment (SOFA) (3).

Raschke et al. (1) evaluated the performance of the APACHE IVa score in subgroups of ICU patients. APACHE is a severity-of-illness score initially created in the 1980s and subsequently updated in 2006 (4,5). This index was developed using data from 110,558 patients from 45 hospitals located throughout the United States, and encompassed 104 intensive care units (ICUs) including mixed medical-surgical, coronary, surgical, cardiothoracic, medical, neurologic, and trauma units. The final model used 142 variables including information from the patient’s medical history, the admission diagnosis, and physiologic data obtained during the first day of ICU admission (4). Although it subsequently has been validated using other large general ICU patient cohorts, its accuracy in subgroups of ICU patients is less clear (6).

To benchmark whether the APACHE IVa performed sufficiently, Raschke et al. (1) employed an interesting and logical strategy. They created a two-variable severity score (2VSS) to define a lower limit of acceptable performance.  As opposed to the 142 variables used in APACHE IVa, the 2VSS used only two variables: patient age and need for mechanical ventilation. They included 66,821 patients in their analysis, encompassing patients from a variety of ICUs located in the southwest United States. The APACHE IVa and 2VSS was calculated for all patients. Although the APACHE IVa outperformed the 2VSS in the general cohort of ICU patients, when patients were divided into subgroups based on admission diagnosis the APACHE IVa showed surprising deficiencies. In patients admitted for coronary artery bypass grafting (CABG), the APACHE IVa did no better in predicting mortality than the 2VSS. The ability of APACHE IVa to predict mortality was significantly reduced in patients admitted for gastrointestinal bleed, sepsis, and respiratory failure as compared to its ability to predict mortality in the general cohort (1).

The work by Raschke et al. (1) convincingly shows that APACHE IVa underperforms when evaluating outcomes in subgroups of patients. In some instances, it did no better than a metric that used only two input variables. But why does this matter? One might argue that the APACHE system was not created to function in this capacity. It was designed and validated using aggregate data. It was not designed to determine prognosis on individual-level patients, or even on subsets of patients. However, in real-world practice it is used to estimate performance in individual ICUs, which have unique cases mixes of patients that may not approximate the populations used to create and validate APACHE IVa. Indeed, other studies have shown that the APACHE IVa yields different performance assessments in different ICUs depending on varying case mixes (2).

So where do we go from here? The work by Raschke et al. (1) is helpful because it offers the 2VSS as an objective method of defining a lower limit of acceptable performance. In the future, more sophisticated and personalized tools will need to be developed to more accurately benchmark ICU quality and performance.  Interesting work is being done using local data to customize outcome prediction (7,8). Other researchers have employed machine learning techniques to iteratively improve predictive capabilities of outcome measures (9,10). As with many aspects of modern medicine, the complexity of severity scoring will likely increase as computational methods allow for increased personalization. Given the importance of accurately assessing quality of care, improving severity scoring will be critical to providing optimal patient care.

Sarah K. Medrek, MD

University of New Mexico

Albuquerque, NM USA

References

1. Raschke RA GR, Ramos KS, Fallon M, Curry SC. The explained variance and discriminant accuracy of APACHE IVa severity scoring in specific subgroups of ICU patients. Southwest J Pulm Crit Care. 2018;17:153-64. [CrossRef]
2. Kramer AA, Higgins TL, Zimmerman JE. Comparing observed and predicted mortality among ICUs using different prognostic systems: why do performance assessments differ? Crit Care Med. 2015;43:261-9. [CrossRef] [PubMed]
3. Vincent JL, Moreno R. Clinical review: scoring systems in the critically ill. Crit Care. 2010;14:207. [CrossRef] [PubMed]
4. Zimmerman JE, Kramer AA, McNair DS, Malila FM. Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today's critically ill patients. Crit Care Med. 2006;34:1297-1310. [CrossRef] [PubMed]
5. Zimmerman JE, Kramer AA, McNair DS, Malila FM, Shaffer VL. Intensive care unit length of stay: Benchmarking based on Acute Physiology and Chronic Health Evaluation (APACHE) IV. Crit Care Med. 2006;34:2517-29. [CrossRef] [PubMed]
6. Salluh JI, Soares M. ICU severity of illness scores: APACHE, SAPS and MPM. Curr Opin Crit Care. 2014;20:557-65. [CrossRef] [PubMed]
7. Lee J, Maslove DM. Customization of a Severity of Illness Score Using Local Electronic Medical Record Data. J Intensive Care Med. 2017;32:38-47. [CrossRef] [PubMed]
8. Lee J, Maslove DM, Dubin JA. Personalized mortality prediction driven by electronic medical data and a patient similarity metric. PLoS One. 2015;10:e0127428. [CrossRef] [PubMed]
9. Awad A, Bader-El-Den M, McNicholas J, Briggs J. Early hospital mortality prediction of intensive care unit patients using an ensemble learning approach. Int J Med Inform. 2017;108:185-95. [CrossRef] [PubMed]
10. Pirracchio R, Petersen ML, Carone M, Rigon MR, Chevret S, van der Laan MJ. Mortality prediction in intensive care units with the Super ICU Learner Algorithm (SICULA): a population-based study. Lancet Respir Med. 2015;3:42-52. [CrossRef] [PubMed]

Cite as: Medrek SK. The need for improved ICU severity scoring. Southwest J Pulm Crit Care. 2019;18:26-8. doi: https://doi.org/10.13175/swjpcc004-19 PDF

Remembering the 100,000 Lives Campaign

Earlier this week the Institute for Healthcare Improvement (IHI) emailed its weekly bulletin celebrating that it has been ten years since the end of the 100,000 Lives Campaign (Appendix 1). This was the campaign, according to the bulletin, that put IHI on the map. The Campaign started at the IHI National Forum in December 2004, when IHI's president, Don Berwick, announced that IHI would work together with nearly three-quarters of the US hospitals to reduce needless deaths by 100,000 over 18 months. A phrase borrowed from political campaigns became IHI's cri de coeur: “Some is not a number. Soon is not a time.”

The Campaign relied on six key interventions:

• Rapid Response Teams
• Improved Care for Acute Myocardial Infarction
• Medication Reconciliation
• Preventing Central Line Infections
• Preventing Surgical Site Infections
• Preventing Ventilator-Associated Pnemonia [sic]

According to the bulletin, the Campaign’s impact rippled across the organization and the world. IHI listed some of the lasting impacts:

• IHI followed with the 5 Million Lives Campaign – a campaign to avoid 5 million instances of harm.
• Don Berwick and Joe McCannon brought lessons from leading the Campaigns to Centers for Medicare and Medicaid Services (CMS) and the Partnership for Patients.
• Related campaigns were launched in Canada, Australia, Sweden, Denmark, UK, Japan, and elsewhere.

IHI's profile definitely grew. One indicator tracked by IHI was media impressions, which rose to 250 million in the final year of the Campaign. IHI even put a recreational vehicle on the streets to promote their Campaign (Appendix 1). Campaign Manager Joe McCannon was on CNN to discuss the results of the Campaign.

How did IHI achieve such remarkable results in saving patients' lives? The answer is they did not. Review of the evidence basis for at least 3 of these interventions revealed fundamental flaws (1). The largest trial of rapid response teams failed to result in any improvements and the interventions to prevent central line infections and ventilator-associated pneumonia were non- or weakly-evidenced based and unlikely to improve patient outcomes (2-4). The poor methodology and sloppy estimation of the number of lives saved were pointed out in the Joint Commission’s Journal of Quality and Safety by Wachter and Pronovost (5). IHI failed to adjust their estimates of lives saved for case-mix which accounted for nearly three out of four "lives saved." The actual mortality data were supplied to the IHI by hospitals without audit, and 14% of the hospitals submitted no data at all. Moreover, the reports from even those hospitals that did submit data were usually incomplete. The most striking example is that the IHI was so anxious to announce their success that the data was based on only 15 months of data. The final three months were extrapolated from hospitals’ previous submissions. Important confounders such as the background of declining inpatient mortality rates were ignored. Even if the Campaign "saved" lives, it would be unclear if the Campaign had anything to do with the reduction (5). Buoyed by their success, the IHI proceeded with the 5,000,000 Lives Campaign (6). However, this campaign ended in 2008 and was apparently not successful (7). Although IHI promised to publish results in major medical journals, to date no publication is evident.

A fundamental flaw in the logic behind the 100,000 Lives Campaign was that preventing a complication, for example an infection, results in a life saved. Many of our patients in the ICU have an infection as their life-ending event. However, the patients are often in the ICU because their underlying disease(s). In many instances their underlying disease(s) such as cancer, heart disease, or chronic obstructive pulmonary disease are so severe that survival is unlikely. It is akin to poisoning, stabbing, shooting and decapitating a hapless victim and saying that had the decapitation been prevented, survival was assured. IHI also assumed that the data was collected completely and honestly. However, the data was incomplete as pointed out above and the honesty of self-reported hospital data has also been called into question (8).

The bulletin correctly pointed out that Berwick did carry this political campaign with its sloppy science to Washington as CMS' administrator. Under Berwick's leadership, CMS would announce a campaign, have the hospitals collect the data, extrapolate the mortality or other benefit, and prepare a press release. This scheme continues until this day (9). CMS further confounded the data by providing financial incentives to hospitals, often resulting in bonuses to hospital executives, making the data further suspect. Certainly, CMS would not examine the hospital data with skepticism because the success of their campaign was in their own political best interest.

The 100,000 Lives Campaign also had one other outcome. It made many of us who believe in the power of evidence-based medicine to enrich patients' lives to be suspicious of these political maneuvers. To rephrase a well-known quote, "The first victim of politics is the truth". These campaigns certainly financially benefit hospitals and their administrators and politically benefit bureaucrats, but whether they benefit patients is questionable. The bulletin from IHI should be viewed for what it is, a political self-promotion to rewrite the failed history of the 100,000 Lives Campaign.

Richard A. Robbins, MD

Editor, SWJPCC

References

1. Robbins RA. The unfulfilled promise of the quality movement. Southwest J Pulm Crit Care. 2014;8(1):50-63. [CrossRef]
2. Hillman K, Chen J, Cretikos M, Bellomo R, Brown D, Doig G, Finfer S, Flabouris A; MERIT study investigators. Introduction of the medical emergency team (MET) system: a cluster-randomised controlled trial. Lancet. 2005;365(9477):2091-7. [CrossRef] [PubMed]
3. Hurley J, Garciaorr R, Luedy H, Jivcu C, Wissa E, Jewell J, Whiting T, Gerkin R, Singarajah CU, Robbins RA. Correlation of compliance with central line associated blood stream infection guidelines and outcomes: a review of the evidence. Southwest J Pulm Crit Care 2012;4:163-73.
4. Padrnos L, Bui T, Pattee JJ, Whitmore EJ, Iqbal M, Lee S, Singarajah CU, Robbins RA. Analysis of overall level of evidence behind the Institute of Healthcare Improvement ventilator-associated pneumonia guidelines. Southwest J Pulm Crit Care 2011;3:40-8.
5. Wachter RM, Pronovost PJ. The 100,000 Lives Campaign: A scientific and policy review. Jt Comm J Qual Patient Saf. 2006;32(11):621-7. [PubMed]
6. Institute for Healthcare Improvement. 5 million lives campaign. Available at: http://www.ihi.org/about/Documents/5MillionLivesCampaignCaseStatement.pdf (accessed 6/24/16).
7. DerGurahian J. IHI unsure about impact of 5 Million campaign. Available at: http://www.modernhealthcare.com/article/20081210/NEWS/312109976 (accessed 6/24/16).
8. Meddings JA, Reichert H, Rogers MA, Saint S, Stephansky J, McMahon LF. Effect of nonpayment for hospital-acquired, catheter-associated urinary tract infection: a statewide analysis. Ann Intern Med. 2012;157:305-12. [CrossRef] [PubMed]
9. AHRQ Report: Hospital-Acquired Conditions Continue To Decline, Saving Lives and Costs. Dec 1, 2015. Available at: http://www.ahrq.gov/news/newsletters/e-newsletter/496.html#1 (accessed 6/24/16).

Cite as: Robbins RA. Remembering the 100,000 lives campaign. Southwest J Pulm Crit Care. 2016;12(6):255-7. doi: http://dx.doi.org/10.13175/swjpcc058-16 PDF 

CMS Penalizes 758 Hospitals for Safety Incidents

The Centers for Medicare and Medicaid Services (CMS) is penalizing 758 hospitals with higher rates of patient safety incidents, and more than half of those were also fined last year, as reported by Kaiser Health News (1).

Among the hospitals being financially punished are some well-known institutions, including Yale New Haven Hospital, Medstar Washington Hospital Center in DC, Grady Memorial Hospital, Northwestern Memorial Hospital in Chicago, Indiana University Health,  Brigham and Womens Hospital, Tufts Medical Center, University of North Carolina Hospital, the Cleveland Clinic, Hospital of the University of Pennsylvania, Parkland Health and Hospital, and the University of Virginia Medical Center (Complete List of Hospitals Penalized 2016). In the Southwest the list includes Banner University Medical Center in Tucson, Ronald Reagan UCLA Medical Center, Stanford Health Care, Denver Health Medical Center and the University of New Mexico Medical Center (for list of Southwest hospitals see Appendix 1). In total, CMS estimates the penalties will cost hospitals $364 million. Look now if you must, but you might want to read the below before on how to interpret the data.

The penalties, created by the 2010 health law, are the toughest sanctions CMS has taken on hospital safety. Patient safety advocates worry the fines are not large enough to alter hospital behavior and that they only examine a small portion of the types of mistakes that take place. On the other hand, hospitals say the penalties are counterproductive and unfairly levied against places that have made progress in safety but have not caught up to most facilities. They are also bothered that the health law requires CMS to punish a quarter of hospitals each year. CMS plans to add more types of conditions in future years.

I would like to raise two additional concerns. First, is the data accurate? The data is self-reported by the hospitals and previously the accuracy of these self reports has been questioned (2). Are some hospitals being punished for accurately reporting data while others rewarded for lying? I doubt that CMS will be looking too closely since bad data would invalidate their claims that they are improving hospital safety. It seems unlikely that punishing half the Nation's hospitals will do much except encouraging more suspect data.

Second, does the data mean anything? Please do not misconstrue or twist the truth that I am advocating against patient safety. What I am advocating for is meaningful measures. Previous research has suggested that the measures chosen by CMS have no correlation or even a negative correlation with patient outcomes (3,4). In other words, doing well on a safety measure was associated with either no improvement or a negative outcome, in some cases even death. How can this be? Let me draw an analogy of hospital admissions. About 1% of the 35 million or so patients admitted to hospitals in the US die. The death rate is much lower in the population not admitted to the hospital. According to CMS' logic, if we were to reduce admissions by 5% or 1.75 million, 17,500 lives (1% of 1.75 million) would be saved. This is, of course, absurd.

Looking at hospital acquired infections which make up much of CMS' data, CMS' logic appears similar. For example, insertion of urinary catheters, large bore central lines or endotracheal intubation in sick patients is common. The downside is some will develop urinary, line or lung infections as a complication of these insertions. Many of these sick patients will die and many will have line infections. The data is usually reported by saying hospital-acquired infections have decreased saving 50,000 lives and saved $12 billion in care costs (5). However, the truth is that hospital-acquired infections are often either not the cause of death or the final event in a disease process that caused the patient to be admitted to the hospital in the first place. If 50,000 lives are saved that should be reflected in the hospital death rates or a savings on insurance premiums. Neither has been shown to my knowledge.

So look at the data if you must but look with a skeptical eye. Until CMS convincingly demonstrates that the data is accurate and that their incentives decrease in-hospital complications, mortality and costs-the data is suspect. It could be as simple that the hospitals receiving the penalties are those taking care of sicker patients. What this means is that some hospitals, perhaps the ones that need the money the most, will have 1% less CMS reimbursement, which might make care worse rather than better.

Richard A. Robbins, MD

Editor

SWJPCC

References

1. Rau J. Medicare penalizes 758 hospitals for safety incidents, Kaiser Health News. December 10, 2015. Available at: http://khn.org/news/medicare-penalizes-758-hospitals-for-safety-incidents/ (accessed 12/11/15).
2. Robbins RA. The Emperor has no clothes: the accuracy of hospital performance data. Southwest J Pulm Crit Care 2012;5:203-5.
3. Robbins RA, Gerkin RD. Comparisons between Medicare mortality, morbidity, readmission and complications. Southwest J Pulm Crit Care. 2013;6(6):278-86
4. Lee GM, Kleinman K, Soumerai SB, et al. Effect of nonpayment for preventable infections in U.S. hospitals. N Engl J Med. 2012;367(15):1428-37. [CrossRef] [PubMed]
5. Department of Health and Human Services. Efforts to improve patient safety result in 1.3 million fewer patient harms, 50,000 lives saved and $12 billion in health spending avoided. December 2, 2014. Available at: http://www.hhs.gov/about/news/2014/12/02/efforts-improve-patient-safety-result-1-3-million-fewer-patient-harms-50000-lives-saved-and-12-billion-in-health-spending-avoided.html (accessed 12/11/15).

Cite as: Robbins RA. CMS penalizes 758 hospitals for safety incidents. Southwest J Pulm Crit Care. 2015;11(6):269-70. doi: http://dx.doi.org/10.13175/swjpcc153-15 PDF

Smoking, Epidemiology and E-Cigarettes

"The true face of smoking is disease, death and horror - not the glamour and sophistication the pushers in the tobacco industry try to portray." - David Byrne

In our fellows’ conference we recently reviewed the evolution of the science of clinical epidemiology as it relates to the association of smoking and lung cancer and the concurrent history of tobacco marketing in the United States. 

This story begins in 1950, when Richard Doll and Austin Bradford Hill published their landmark case control study demonstrating the association between smoking and lung cancer (1). This study was performed with methodological standards that have rarely been matched in the 63 years since.  Exhaustive analysis of possible confounders, a multi-stage evaluation of study blinding, determination of dose-effect, and the use of multiple analyses to establish consistency are among many examples of superb attention to detail exercised by Doll and Hill in this study.  The results showed that patients with lung cancer were about 15 times more likely than matched control patients to have smoked tobacco (Odds ratio 15).  The p-value was 0.00000064  - indicating that the probability of calculating such a result by chance alone is less than one-in-a-million.  In comparison, many modern case control trials are characterized by weak associations (odds ratios of 1-3) with p-values that are barely significant.  Yet the phenomenal and nearly unparalleled results of this study had practically no discernable effect on the increasing rate of smoking in the following decade.

Many factors opposed the conclusions of Doll and Hill.  Atmospheric pollution – perhaps emanating from motor car exhaust or asphalt tarmac – was felt to be the leading suspect in the increasing incidence of lung cancer.  At the time, it seemed inconceivable to most people that smoking could cause cancer.  Two thirds of British men smoked.  Smoking was widely endorsed by the medical profession – Doll and Hill themselves had both previously been smokers.  The British Department of Health did not endorse their findings, amid worries that the study might start a panic.  Several prominent statisticians, including Sir Ronald Fisher, publicly criticized their study design and conclusions.  Fisher was a polymath – a genius with significant accomplishments in multiple disciplines, widely recognized as the founder of modern statistics, having invented Fisher’s exact test, and ANOVA and having collaborated in the development of the Student’s T test.  Fisher was also an avid smoker.  It was later disclosed that Fisher had lucrative financial ties to the tobacco industry, raising questions whether Fisher’s criticisms of Doll and Hill were bought and paid for.

Doll and Hill followed up with a stronger study design – performing one of the finest cohort studies ever – the British Physician’s study.  They enrolled over 40,000 British Physicians – almost 70% of all registered in Britain.  Outcomes in this cohort were eventually evaluated over 50 years, and contributed to our knowledge in many areas of medicine.  But the results in regards to the relationship between smoking and lung cancer were objectively convincing within the first decade of follow-up.  In an interim analysis in 1961 (2), the relative risk for lung cancer in smokers was found to be increased 18 times – consistent with the findings of their case control trial.  Fisher’s exact test was incalculable in 1961 since it required the quantization of enormous factorials, but I calculated a p-value of 0.0000000000000001 (one in 100-quadrillion) using their data and an on-line Microsoft statistics program.  It’s satisfying to find that Fisher’s namesake statistic so convincingly validates the conclusions that he personally refuted.  Sir Austin Bradford Hill is famous for his contention that we often over-focus on achieving a p-value < 0.05 in modern medical research – the incomparable statistical significance of this study illustrates his point.  

Despite increasing scientific evidence against smoking, cigarette consumption in the U.S. continued to rise, and did not fall below pre-1950 levels until the early eighties.  A further generation of young men took on the habit, many of which were introduced to smoking in the armed services - cigarettes having been routinely included in C-rations of US soldiers who fought in WWII, Korea and Viet Nam.  Cigarette smoking was endorsed by everyone from movie stars, to sports stars to doctors – Bob Hope, Mickey Mantle and Ronald Reagan among them.  Santa Claus appeared in multiple ads with a cigarette in one hand, and his red toy bag in the other – fecklessly endorsing multiple different brands including Lucky Strikes and Pall Malls. 

Several tobacco advertisement campaigns were particularly influential.  Philip Morris introduced the “Marlboro Man”, considered one of the most brilliant ad campaigns in history, in 1954.  Marlboro cigarettes were filtered.  The implied (but factitious) protective benefits of the filter were not explicitly marketed, but filtered cigarettes were considered “feminine” at the time.  The use of real rodeo cowboys in the Marlboro ads dramatically changed that impression – particularly in the minds of post adolescent boys.  One indication of the success of the Marlboro Man is that Philip Morris is said to have spent $300 million dollars finding a replacement when Darrell Winfield, the most famous of the Marlboro men, retired.

In the late sixties, Philip Morris also marketed smoking to young women with a brand designed specifically for women called Virginia Slims.  Riding the wave of women’s liberation, the slogan “You’ve come a long way baby” promoted smoking as a way to express emancipation and empowerment.   RJ Reynolds introduced the “Joe Camel” ad campaign in 1987, allegedly targeting children with a cool-looking cartoon of an anthropomorphic camel.  Sounds silly, I know, but it worked.  In 5 short years after starting this campaign, the annual sales of Camel cigarettes to teenagers rose from 6 million to 470 million dollars.  At its peak, it was shown that six-year-old children could associate the character of “Joe Camel” with Camel cigarettes about as frequently as they could associate Mickey Mouse with Disney.  A study published in JAMA concluded that tobacco experimentation by 700,000 adolescents per year could be attributed to targeted advertising (3). 

Although public education had already made great inroads in reducing smoking in the US by the 80’s, legal and governmental anti-smoking pressure began to build thereafter.  In 1988, Rose Cipollone  posthumously won the first successful wrongful harm lawsuit of a smoker against a tobacco manufacturer.  Mangini sued RJ Reynolds on behalf of children in regards to the Joe Camel ad campaign.  In the 1988 Report of the Surgeon General, C Everett Koop concluded that nicotine has an addictiveness similar to that of heroin.  C Everett Koop’s continuing efforts to raise public awareness initiated some of the first public discourse in regards to the dangers of second-hand smoke (subsequently found to cause 50,000 deaths per year in the U.S.).  Smoking rates in the United States declined from 38% to 27% during his tenure.

In the 1990s, the tobacco lobby engaged in a comprehensive and aggressive political effort to neutralize clean indoor air legislation, minimize tobacco tax increases, and preserve the industry's marketing strategies.  However the famous Waxman congressional hearings intervened in 1997.  In sworn testimony before congress, the CEOs of seven major tobacco companies famously asserted that smoking tobacco was not addictive, contrary to incontrovertible scientific evidence.  Two sources revealed their insincerity.  The first was testimony of previous employees of the tobacco industry, such as Jeffrey Wigman and Victor DeNoble, who testified that the addictive and carcinogenic properties of cigarette tobacco had been artificially manipulated by the industry.   The second was the discovery of internal tobacco industry memos, which revealed that the addictive properties of tobacco were well recognized within the industry as early as 1960s.  A few excerpts follow:

“… nicotine is addictive. We are, then, in the business of selling nicotine, an addictive drug” July 17, 1963 report by then Brown & Williamson general counsel/vice president Addison Yeaman.

 “The cigarette should be conceived not as a product but as a package. The product is nicotine. …Think of a cigarette as a dispenser for a dose unit of nicotine…”  1972 William Dunn, Jr., of the Philip Morris Research Center, “Motives and Incentives in Cigarette Smoking.”

“Within 10 seconds of starting to smoke, nicotine is available in the brain. . . giving an instantaneous catch or hit . . . Other “drugs” such as marijuana, amphetamines, and alcohol are slower”  Circa 1980  C.C. Greig in a BAT R&D memo

The Waxman hearings resulted in a $368 billion dollar assessment against the tobacco industry, and increased restrictions on advertising and lobbying.  Shortly thereafter, the Joe Camel and Marlboro Man ad campaigns were terminated.  With the public revelation that three previous Marlboro Men had died from lung cancer, that ad campaign had lost its appeal.  

In the late 90s/early 2000s, the nicotine content of all major brands of cigarettes was progressively increased on average by 1.8% per year.  This might theoretically make it harder for smokers to kick the habit.  Sales promotions totaling about $400 per year per smoker were directed at loyal smokers.  Despite restrictions, the tobacco industry continued to invest $25 million dollars per year in lobbying.  Upon further negotiation, the tobacco master settlement was reduced to 200 billion – only 12.7 billion to be paid up front.  The full details of this settlement have become increasingly legally obfuscated over time in my opinion; some states are actually selling tobacco settlement bonds now to protect themselves against loss of future return from the settlement.   

Although US cigarette consumption has dramatically fallen, worldwide sales are peaking, and the international rates of women smokers are still on the rise.  Philips Morris restructured and rebranded their corporation as Altria (sounds like the word “altruistic”).  They subsumed Kraft and Nabisco foods, but the majority of their >100,000 million dollars in annual revenue are derived from tobacco sales, about two-thirds of which are international. 

Many US tobacco firms are rapidly investing in production and marketing of electronic cigarettes that vaporize nicotine for inhalation.  It is likely that inhaling vaporized nicotine is less dangerous than smoking tobacco.  However, the health effects of inhaling vaporized nicotine are not well studied yet.  The purported benefits of vaping over smoking have already been publicly aired as an argument to turn back current restrictions on public smoking.  Electronic cigarettes are being advertised as glamorous again in advertisements reminiscent of tobacco ads seen in the 1970s.  E-cigs in which nicotine is flavored with chocolate, or various fruit flavors, seem to once-again target children.  The promotion of a highly addictive drug to children and young adults cannot be beneficial to society in the long term, even if vaping doesn’t lead to lung cancer.  But the rapid increase in vaping promises that another round in the societal struggle against nicotine addiction is about to begin again.

Doll and Hill’s work played a tremendous beneficial role in this story.  Their case control and cohort studies set the methodological standard by which all subsequent observational trials should be measured – although our experience in journal club is that modern observational trials don’t even come close.  Furthermore, their work became the basis for the subsequent formulation of the “Bradford Hill Criteria” for establishing causation, which still plays a dominant role in medical and medicolegal reasoning.   

Robert A. Raschke, MD

Associate Editor 

References

1. Doll R, Hill AB. Smoking and carcinoma of the lung; preliminary report. Br Med J. 1950;2(4682):739-48. [CrossRef]
2. Doll R, Hill AB. The mortality of doctors in relation to their smoking habits; a preliminary report. Br Med J. 1954;1(4877):1451-5. [CrossRef]
3. Pierce JP, Choi WS, Gilpin EA, Farkas AJ, Berry CC. Tobacco industry promotion of cigarettes and adolescent smoking. JAMA. 1998;279(7):511-5. [CrossRef] [PubMed]   

Reference as: Raschke RA. Smoking, epidemiology and e-cigarettes. Southwest J Pulm Crit Care. 2013;7(1):41-5. doi: http://dx.doi.org/10.13175/swjpcc092-13 PDF