Intended for healthcare professionals


Patients need safer hospitals, every day of the week

BMJ 2015; 350 doi: (Published 15 April 2015) Cite this as: BMJ 2015;350:h1826
  1. Kumar Dharmarajan, assistant professor of medicine1,
  2. Nancy Kim, assistant professor of medicine2,
  3. Harlan M Krumholz, professor of medicine1
  1. 1Section of Cardiovascular Medicine, Yale University School of Medicine, New Haven, CT 06510, USA
  2. 2Section of General Internal Medicine, Yale University School of Medicine, New Haven, CT, USA
  1. Correspondence to: K Dharmarajan kumar.dharmarajan{at}

Hospital acquired conditions must be reduced throughout the week, not just at weekends

In the United States, efforts to improve patient safety have focused, in part, on reducing hospital acquired conditions, or HACs. Hospital acquired conditions include healthcare associated infections, iatrogenic complications, and other adverse events occurring during treatment for another condition. Historically, the additional costs incurred by these adverse events led to greater payments by the government.1 To encourage a reduction in these potentially preventable conditions, the US Centers for Medicare and Medicaid Services (CMS) in 2008 decided not to pay for the higher costs of care resulting from 14 specific hospital acquired conditions.2 Starting this year, the CMS’s HAC Reduction Program will levy considerable financial penalties on hospitals scoring poorly on a composite measure of performance regarding hospital acquired conditions, which is derived from a review of administrative claims and medical record data.3 4

In a linked paper, Attenello and colleagues (doi:10.1136/bmj.h1460) determined whether weekend admission is associated with a greater likelihood of developing these adverse events,5 as weekend admission has been associated with worse outcomes for a large number of life threatening conditions, such as myocardial infarction,6 pulmonary embolism,7 and intracerebral haemorrhage.8 They examined 2002-10 data from the National Inpatient Sample, the largest publicly available, all payer inpatient database of administrative claims in the United States that enables the calculation of population level estimates of the occurrence of hospital acquired conditions. The authors identified discharge diagnosis codes for the presence of 14 such conditions, including advanced pressure ulcers, falls and trauma, catheter associated infections, and other adverse events. They then examined whether weekend admission to hospital was associated with a greater likelihood of acquiring a hospital acquired condition after controlling for various patient and hospital characteristics within the administrative data.

The authors found that hospital acquired conditions were two absolute percentage points more likely among people admitted at weekends (5.7% of admissions) than among those admitted on weekdays (3.7%). The authors also provided comparative estimates of the frequency of a broad spectrum of hospital acquired conditions and found that falls and associated trauma comprised more than 12.6 million of the 14.3 million reported conditions within the dataset.

The study has considerable limitations inherent to the data source. As the authors recognise, weekend admissions are more complex than weekday admissions. Few elective admissions occur at the weekend, and the conditions prompting admission are different.7 9 As Attenello and colleagues note, patients admitted at weekends more often have impaired functional status. Unfortunately, the National Inpatient Sample does not contain detailed information collected by clinicians to help identify and explore further the potential differences in case mix between weekend and weekday admissions. Moreover, as ascertainment of hospital acquired conditions depended on administrative codes, it was impossible to know when during the admission each condition occurred and whether it was related to the weekend.

It is therefore too early to suggest that interventions specifically targeting care at weekends would lower the frequency of hospital acquired conditions. Patients are more likely to benefit from an approach analogous to universal precautions, with systems that reduce everyday risk. Even among weekday admissions, the rates of hospital acquired conditions are unacceptably high. There are good reasons to introduce systems for a safer environment every day of the week.

Nevertheless, the linked paper highlights important challenges faced by researchers studying these adverse events and trying to measure their occurrence. Limited fields within administrative data make it hard to understand the reasons behind many hospital acquired conditions, as both patient and care related factors are poorly recorded. In addition, the accuracy of estimates derived from administrative codes has not been convincingly validated against medical record data for many of the patient safety indicators for hospital acquired conditions used to measure hospitals performance.4 And even when incidence estimates are accurate, administrative data give little indication of the severity of these events, including their impact on patient outcomes.

The study by Attenello and colleagues also highlights problems with how measures for the occurrence of hospital acquired conditions are being deployed. For example, the HAC Reduction Program focuses on the small subset of traumatic falls resulting in postoperative hip fracture rather than the much larger group of patients with falls described in the new study. In addition, some hospitals will be allowed to report their performance selectively, using measures of their choosing,3 such as those for catheter associated urinary tract infections and central line associated bloodstream infections. Both are relatively under-reported in the current study10 11 and would benefit from improved national estimates.

Finally, measures are not harmonised across different regulatory bodies. The 14 hospital acquired conditions targeted by the CMS in 2008 are not the same ones measured by the HAC Reduction Program or the CMS Hospital Value-Based Purchasing Program, both of which direct financial penalties towards hospitals with higher than expected rates of hospital acquired conditions.4

Attenello and colleagues’ study provides a service by raising questions about the timing of the occurrence of hospital acquired conditions, and it illustrates important challenges in the measurement and prevention of these adverse events. Although the authors found that patients admitted at weekends were more likely to acquire a hospital acquired condition, work is needed to determine when during the week these excess conditions occur and how they relate to the severity of illness and hospital care. In the absence of further data, hospitals should focus on the reduction of hospital acquired conditions more generally, not just at weekends. A universal approach is needed to make hospitals safer.


Cite this as: BMJ 2015 350:h1826


  • Research, doi:10.1136/bmj.h1460
  • Competing interests: We have read and understood the BMJ policy on declaration of interests and declare the following: KD, NK, and HMK work under contract with the Centers for Medicare and Medicaid Services in the United States to develop and maintain performance measures; HMK is chair of a cardiac scientific advisory board for UnitedHealth; HMK is the recipient of research grants from Medtronic and Johnson & Johnson through Yale University.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription