The road to patient hell is paved with quality metrics: breastfeeding edition

Road to hell

Why do hospitals let bad things happen to good people?

The reason can often be found in efforts to meet government quality metrics. As a recent editorial in JAMA, Unintended Harm Associated With the Hospital Readmissions Reduction Program, explains:

… The HRRP [Hospital Readmissions Reduction Program] imposed financial penalties on hospitals based on rates of 30-day risk-standardized hospital readmission for heart failure, acute myocardial infarction, and pneumonia, with up to 3% of a hospital’s total Medicare revenue from admissions for any condition (target or nontarget) at risk. In fiscal year 2018, 80% of the hospitals subject to the HRRP have been penalized, amounting to $564 million in reduced payments by Medicare.

The government wanted to improve the quality of care of those who had to be readmitted ostensibly because they hadn’t been adequately treated during the initial hospitalization.

Many babies are starving and some are starving to death.

And it worked! Hospital readmissions were reduced and payments to hospitals were also reduced.

There was just one teeny, tiny problem:

Using an interrupted time-series analysis approach, an analysis of clinical data from Get With The Guidelines – Heart Failure linked to Medicare data demonstrated an increase in 30-day and 1-year mortality associated with the implementation of the HRRP…

The analysis also demonstrated that the overall increase in mortality associated with the HRRP was mainly driven by patients who were not readmitted to the hospital, but who died within 30 days of discharge. This finding, in particular, enhances the likelihood of a causal relationship between the HRRP financially incentivized restricting of inpatient readmissions and the harm observed.

The “successful” program killed patients by restricting access to the medical care they needed to survive.

How did things go so very wrong? It’s all about the assumptions:

Assumption 1: Cardiac patients readmitted within 30 days would not have been readmitted if they had been better treated in the initial admission. In other words, readmission was a sign of low quality.
Assumption 2: Refusing to pay for readmissions would incentivize hospitals and providers to improve quality of the initial admission.
Assumption 3: Patient health would be improved while money was being saved.

All three assumptions turned out to be wrong and a modicum of understanding of both medical care and human nature would have made that it clear that the deadly results were practically inevitable.

1. Readmission rates are not a quality metric in this setting of frail, elderly people suffering chronic diseases. Readmission is a function of how sick the patients are not how good the treatment is.
2. Since readmission rates are not a quality metric, no amount of incentivization can improve the health of discharged patients.
3. By penalizing hospitals for relapses they could not prevent, they incentivized hospitals to refuse to readmit the very patients that needed readmission most. Those patients died.

This is what happens when government (or any institution) chooses to make a metric that has little to do with quality of care into a quality metric and penalizes hospitals and providers who can’t reach it.

The same thing has happened to breastfeeding support with the same deadly result, although in this case the patients who are harmed are newborn babies and their mothers.

Contemporary breastfeeding promotions efforts, in particularly the Baby Friendly Hospital Iniative (BFHI), have made breastfeeding rates into a quality metric by employing the same faulty assumptions that let to the deaths of heart patients.

Assumption 1: All women can and would breastfeed exclusively if they received enough education and support.
Assumption 2: Incentivizing hospitals to improve breastfeeding rates will lead to more babies being breastfed successfully.
Assumption 3: Infant health will be improved.

Aggressive breastfeeding promotion efforts have also “worked.” Exclusive breastfeeding rates on discharge have risen by more than 200%.

At the same time newborn hospital readmissions for breastfed babies are double that of bottle fed babies accounting for tens of thousands of hospital readmissions each year in the US alone. Indeed the single biggest risk factor for newborn hospital readmission is exclusive breastfeeding on discharge. Breastfeeding is a leading factor or the leading factor in an increase in neonatal hypernatremic dehydration, kernicterus (severe jaundice) and the permanent brain injuries and deaths that can result.

Many babies are starving and some are starving to death.

What went wrong? All three assumptions turned out to be wrong and a modicum of understanding of both medical care and human nature would have made that it clear that deadly results were practically inevitable.

1. Breastfeeding rates are NOT a quality metric; not all women can produce enough breastmilk to fully nourish a baby and not all women want to breastfeed.
2. Since not every mother can or wants to breastfeed, incentivizing hospitals to increase breastfeeding leads them to pressure, shame and punish mothers who wish to use formula.
3. Incentivizing hospitals to improve breastfeeding rates to 100% when approximately 15% of first time mothers cannot produce enough breastmilk in the early days to adequately nourish an infant leads hospitals to withhold formula from babies who desperately need it. Incentivizing hospitals to increase rates of skin-to-skin contact and rooming in leads to an increase in newborn deaths as infants smother in or fall from their mothers’ hospital beds.

These deadly results were nearly inevitable once government decided to make breastfeeding rates — which are not and never were a metric of quality — into a quality metric.

When government makes readmission rates a quality metric, hospitals respond by withholding readmission and patients die as a result. When government makes breastfeeding rates a quality metric, hospitals respond by withholding formula and babies die as a result.

The road to patient hell is paved with quality metrics. In the case of breastfeeding promotion, it’s pretty easy to save babies and mothers from hell: just stop pretending that breastfeeding rates have anything to do with either hospital quality or infant health.


  • andrea

    Thank you many times over!

  • RudyTooty

    thank you.

  • Same goes for cesarean rates. Quality metrics are well-intended but fraught with unintended consequences and a lack of imagination and curiosity about what underlies the observed statistics. They should always have been a starting point – something that triggers an exploration of why, and if that examination determines that a process or way of doing things is wanting, then change the process and observe again. The obsession with the metric as an ends to itself is the problem, as is the assumption that observed reductions mean underlying problems have been “fixed”. It’s hard to listen to patients when their voices have been drowned out by the statistics.

  • BeatriceC

    Tiny typo alert: “Aggressive breastfeeding promotion efforts have also “worked.” Exclusive breastfeeding rates on discharge have rise by more than 200%.” The n has been left off of what I assume should be “risen”.