Displaying 1-10 letters out of 98 published
SPC Versus GAM for hospital adverse events arising in a complex system
Statistical process control works well when there is independence and linearity. Complex systems produce data that are often not independent, often nonlinear and display self-organisation and emergent behaviour. To say that statistical process control works when behaviour is emergent may make little sense. Increasingly adverse events like colonisation with antibiotic-resistant organisms arise in a complex system. Although the latter may be influenced favourably by using bundles, checklists and morbidity and mortality analysis, this is not the same as statistical process control. Each recent year and especially the most recent, we have seen the apparent superiority of generalised additive models in preparing summaries of the hospital's infection management annual report. The idea that generalised additive models work best when there is a complex system could be important if substantiated.
Conflict of Interest:
Improving safety culture to reduce adverse events
Reynolds et al1 reported the impact of providing prescriber feedback in reducing prescribing errors. The authors have concluded that reducing prescribing errors needs a multifaceted approach and feedback alone is not sufficient. Medication errors are often preventable and inappropriate prescribing is identified as an important contributing factor to medication errors.2 It is interesting to note that despite regular feedback, prescribing errors are not improved. It shows a failure in improving underlying prescribing culture.
We need a shift in how we consider safety issues in an organisation. It is important to assess underlying safety climate in an organisation. Medication safety issues should be discussed as part of interdisciplinary rounding or daily safety huddles. If medication errors were considered as a safety concern during huddles or interdisciplinary rounding then prescribers could see how errors can impact individual patients and it may result in practice changes. Interventions that focus on improving safety culture has proven to be effective in reducing adverse events in hospitals such as catheter associated blood stream infections.3
It is also well recognised that a committed leadership and supportive organisational culture is important in bringing practice changes.4 Senior doctors play an important role in developing the prescribing culture of junior doctors. Junior doctors have reported that early in their career, their senior colleagues primarily influence prescribing practices.5 While the authors decision to target junior doctors prescribing is based on valid reasons, the lack of involvement of senior doctors in the feedback process may have limited its effectiveness in bringing practice changes. It may have been worthwhile considering the prescribing practices of a team including the consultants and not just the prescribing of junior doctors.
In conclusion, it is evident that feedback alone does not change prescribing practices and we need a shift in our approach towards safety and build an organisational culture that consider safety as a key priority.
1. Reynolds M, Jheeta S, Benn J, et al. Improving feedback on junior doctors' prescribing errors: mixed-methods evaluation of a quality improvement project. BMJ Qual Saf 2016:bmjqs-2015-004717. 2. Kohn LT, Corrigan JM, Donaldson MS. To err is human: building a safer health system. Washington D.C: National Academies Press, 2000. 3. Pronovost PJ, Berenholtz SM, Goeschel CA, et al. Creating high reliability in health care organizations. Health Serv Res 2006;41(4p2):1599-617. 4. Kaplan GS, Patterson SH, Ching JM, et al. Why Lean doesn't work for everyone. BMJ Qual Saf 2014:bmjqs-2014-003248. 5. De Souza V, MacFarlane A, Murphy AW, et al. A qualitative study of factors influencing antimicrobial prescribing by non-consultant hospital doctors. J Antimicrob Chemother 2006;58(4):840-43.
Conflict of Interest:
The Problem with Root cause analysis
I read with interest the article by Peerally et al (1) on 'The problem with root cause analysis'. I reflected on the recent cases that happened at Royal North Shore Hospital and Sydney Hospital (2,3,4) which led me to consider which investigative tool is best applied to different incidences and identified risks. The use of appropriate tools and involvement of key stakeholders are crucial elements to a successful investigative process and outcomes, however, we cannot ignore the reality of the process cost versus event severity and risk. Use of tools by subject matter expert Root cause analysis (RCA) is a tool used in many investigative incidences (5,6). Often as a result recommendations are made yet similar errors still happen. As correctly mentioned by Peerally et al, most investigations of incidences are done by the local team involved with RCA tools but with a lack of expert accident investigator involvement to ensure regular feedback loops and ongoing corrective actions. I do agree that hospitals should move toward proactively preventing adverse incidences for high probability, high severity risks. Preventing adverse incidences can eliminate harm to patients, reduce liability for organisations and reduce both operating costs and the need for resources. A proactive approach often uses Failure Mode Effect Analysis (FMEA) tools. FMEA often requires a higher level of investigative expertise and as such often costs more so it may be optimal to assess risks on a probability severity matrix to identify which tools are optimal. The proposal of engaging an independent professional body, while preferable, can be time-consuming and expensive. I propose for most cases (with exception for cases with significant legal liability) this level of expertise and independence could be developed within the organisation. The body i.e. quality or risk management department, should comprise of people with qualifications such as system thinking, sound interviewing techniques, able to involve staff, human factor analysis, current clinical practice, health management and have the ability to analyse data (7). This department could then act as a quasi-independent body to avoid situational bias and provide a platform for disseminating the results to intra- hospitals, inter-hospitals and governmental bodies as shared learning to help prevent occurrence or recurrence. As a largely independent department within the organisation, they can for most cases facilitate the investigative processes objectively thus eliminating tendency to blame (8,9,10). Key stake holders' involvement The involvement of key stakeholders is very crucial in any investigative process; leaders, managers, clinicians. The leaders provide governance, leadership and support to the managers. They are involved in the investigative process to gain their input, consensus and to commit resources for any recommendations that might be made. It is critical leaders set departmental performance indicators with due acknowledgment for the resources needed to achieve them as too often the burden of performance and blame is levied on departments, middle management and individuals where identified risk avoidance is under- resourced. The managers (department managers, quality and risk managers) are required to provide a safe environment for practice. They are to ensure that the protocols and standards of care are adhered to and patients are managed in a consistent manner. The role of the manager also includes identifying risks and establish processes to prevent the risks from reaching the patient with the support from the leader. The clinicians are required to conduct the procedures/practices in compliance with their scope of practice, organisational and regulatory boards. Conclusion The usage of an appropriate tool by a qualified person with the right expertise makes a difference. It would be economically unrealistic to apply full FMEA processes for every incident or identified risk profile, so the establishment of an organisational risk severity/probability matrix needs to be developed so the most appropriate tool is used. The involvement of key people ensures that a holistic approach is applied and outcomes of the investigations are implemented with feedback checks and balances and shared across intra-departments, inter-hospitals and at national level (11).
References: 1)Peerally MF, Carr S, Waring J, Dixon-Woods M. The problem with root cause analysis. BMJ Qual Saf. 2016 Aug;1:1-6 2)Bodies swapped: Dead baby mistakenly cremated and daughter finds mother's body mislabelled at Royal North Shore Hospital [television broadcast]. Sydney: The Sydney Morning Herald; 2016 Aug 31. Available from: www.smh.com.au/nsw/daughter-finds-mothers-body-mislabelled-in-morgue -mixup-at-royal-north-shore-hospital-20160830-gr4g3n.html 3)Joseph AP, Hunyor SN. The Royal North Shore Hospital inquiry: a analysis of the recommendations and the implications for quality and safety in Australian public hospitals. Med J Aust. 2008 April ;188(8):469-72 4)Family want justice for fatal gas mix up [television broadcast]. Sydney: Skynews; 2016 Jul 26. Available from: http://www.skynews.com.au/news/top- stories/2016/07/26/incorrect-gas-fitting-behind-nsw-baby-death.html 5)Clifford SP, Mick PB, Derhake, BM. A Case of Transfusion Error in a Trauma Patient with Subsequent Root Cause Analysis Leading to Instituitional Change. J Investig High Impact Case Rep. 2016 May; 4(2):1-4 6)Van-Galen LS, Struik PW, Driesen BEJM, Merten H, Ludikhuize J, Van der Spoel JI, Kramer MHH, Nanayakkara PWB. Delayed Recognition of Deterioration of Patients in General Wards Is Mostly Caused by Human Related Monitoring Failures: A Root Cause Analysis of Unplanned ICU Admissions. 2016 Aug; 11(8):1-14 7) Ibrahim JE. What is the quality of our quality managers? Is it time for quality managers in Australia to be certified? J.Qual Clin Practice. 2000;20(1):32 8)Smetzer JL, Cohen MR. Lessons from the Denver medication error/criminal negligence case: look beyond blaming individuals. Hosp Pharm. 1998;33:640- 57. 9)Leape L. Error in medicine. JAMA. 1994;272:1851-7 10)Runciman W, Merry A, Smith AM. Improving patients' safety by gathering information. Anonymous reporting has an important role. BMJ. 2001;323:7308 11)Leape LL. Why should we report adverse incidents? J Eval Clin Pract. 1999;5:1-4
Conflict of Interest:
A role for patients and the public in improving healthcare
I read with interest the article on what role the patient and public should play in healthcare improvement (1) as this is a question that my organisation has long grappled with and is now required to achieve accreditation against the mandatory National Safety and Quality Health Service Standards (2).
For many years we had a strong Community Advisory Committee and consumers on all key quality and safety committees but it sometimes felt more like we were ticking a box rather than properly engaging with our consumers. We did enjoy some notable successes such as training consumers to participate in our Root Cause Analysis investigation teams and while this garnered national and international interest we didn't feel that we were partnering with our consumers in a meaningful way that acknowledged sufficiently that consumers no longer play a passive role in their healthcare experience.
Following extensive consultation a Partnerships in Care Strategy (3) was developed which examined partnership at the individual (i.e. person centred care), program/department and organisational levels with four key focus areas: Leadership and staff; Empowerment; Feedback and responsiveness; and Environment. Partnering with and empowering our patients and consumers is also one of the six strategic priorities of the organisational strategic plan (4).
The past two years have seen our consumers engaged in many new and exciting ways and also some reinvigoration of existing processes (e.g. evidence based co-design and patient stories). We now have a program that trains managers to interview staff with consumers on the interview panel which has been very successful in the Outpatient and Allied Health setting. Programs/departments are developing strong relationships with the consumers who work with them on quality improvement projects or sit with them on committees and a number are now invited to annual planning days and are considered to be part of the team.
There has also been a large investment in developing electronic tools to capture patient experience information and feedback in the inpatient and outpatient settings which is used to measure the performance of units and inform changes to our processes.
Our ultimate goal is to develop a culture that places patient experience at the heart of everything that we do and where partnering with patients, consumers and carers is the norm (3). Are we there yet? No but the journey is underway and we really are enjoying the ride.
References 1. Ocloo J and Matthews R. From tokenism to empowerment: progressing patient and public involvement in healthcare improvement. BMJ Qual Saf 2016;25: 626-632 2. Australian Commission on Safety and Quality in Health Care. Standard 2: Partnering with Consumers - Safety and Quality Improvement Guide. ACSQHC. 2012 3. Melbourne Health. Partnerships in Care: Working together to improve your experience and outcomes. 2014 4. Melbourne Health. Transforming Health - Melbourne Health Strategic Plan 2015-2020. 2015
Conflict of Interest:
The overuse of thromboprophylaxis in medical patients
In the important editorial of the Grant, is underlined the overuse of thromboprophylaxis in patients hospitalized in the medical field. We agree in emphasizing the difficulty of proper patient assessment that must be carefully evaluated, considering comorbidity and various risk factors,and using the main scores currently in use to assess the start of tromboprofilattica therapy. For this reason we carried out a study where we evaluated 279 patients hospitalized in 21 hospitals in Italy. All patients were negative for the risk of thrombosis to the main risk scales (padua score, Chopard score, Kuscher score). We assessed the frequency of thromboprophylaxis in acutely ill medical patients hospitalized in emergency and internal medicine wards.The results were surprising.Forty-seven patients (16.5%) with negative risk scores were given thromboprophylaxis during hospitalization. On backward stepwise logistic regression analysis, severe infection (odds ratio [OR] 2.31; 95% confidence interval [CI] 1:25 to 4:35) and chronic venous insufficiency (OR 2.3; 95% CI 1.96-4.67) were found to be the strongest predictors of the use of thromboprophylactic treatment with heparin. The subgroup of patients who did not exhibit risk factors was analyzed also, and age was found to be the main factor in the decision- making process Regarding heparin administration in the absence of other risk factors (74.9 ? 11.8 vs 63.7 ? 18.1, p = 0.002). In conclusion, we agree with Grant about the difficulty of identifying patients at moderate risk. We stress also how even patients with low risk then carried out a thromboprophylaxis is not necessary and nd for that reason turns out to be even more 'important to perform a correct stratification of the risk of thromboembolism. Best regards
1 Monti M, Monti A, Bertazzoni G, Pugliese FR, Ciammaichella M, Landolfi R. The overuse of thromboprophylaxis in medical patients: main clinical aspects.G Ital Cardiol (Rome). 2015;16(11):639-43.
Conflict of Interest:
Black box medicine: a structured approach to organisational reflection on major maternal morbidity as a tool for effective organisational learningEditor - Professor Knight(1) highlights a serious problem with systems of organisational learning in maternity care that is endemic across a variety of acute care settings in the NHS. I write to share my experience with a trainee based structured case note review method so other organisations and patients may benefit from what I refer to as a black box medicine (BBM) approach to major maternal morbidity. Trainee based mixed explicit and structured implicit (MESI) retrospective case record review (RCRR) methodology attempts to combine the rigour of external review with resource effectiveness of local review. From personal experience, methodological, logistical and economic barriers often resulted in superficial, subjective and quite unstructured RCRR. Black box medicine evolved from the realisation that organisations needed to improve the RCRR process as learning opportunities were frequently missed. Furthermore, selection of cases for review often focuses on tip of the iceberg phenomenon as resources for a more inclusive review strategy are not available. Consequently patients would continue to be exposed to the same latent suboptimal care. Analysis of care with adverse clinical outcomes in other settings reveals the final common pathway to suboptimal care is failure to recognise and or rescue deteriorating patients. Recommendations for the use of modified obstetric early warning scores(2) reinforce the premise that opportunities to prevent major maternal morbidity lie in the analysis of this final common pathway, and that a BBM approach could enhance organisational learning. Development of MESI RCRR is not a new concept. Recent work has been published on similar RCRR methods(3,4) and the Royal College of Physicians is developing a national RCRR programme to review adult acute care deaths in England and Scotland. However, logistical burden and cost are still significant, while a focus on general adult mortality is probably not applicable to major obstetric morbidity. Development of MESI RCRR that utilises junior members of a MDT to abstract and analyse most of the clinical information would reduce cost per case but threaten validity of the process. Repeated cycles of case note review at various organisations allowed this current method of trainee based RCRR to develop iteratively. Identification of new methodological issues during each cycle allowed refinement based on principles discussed below. Acute care is essentially a series of clinical encounters that can be broadly classified as an assessment, intervention or monitoring event. Every care event has commission or omission characteristics that can be judged as part of a quality assessment process. Commission characteristics include timeliness, appropriateness, sufficiency and absence of adverse event. An event is considered as an omission if it did not occur but was indicated in the clinical context. Good maternal care in any clinical context can therefore be universally defined as an episode consisting of care encounters or events that are timely, appropriate, and adequate without adverse event or omission. Judging care events positively or negatively requires further explicit and structured implicit guidance. Explicit guidance allows decisions on events based on basic physiologic rationale and evidence based standards of care. Structured implicit guidance allows the reviewer to consider medico-legal vulnerability of documentation and the quality of clinical encounters by assessing documented content, evidence of cognitive bias or error, detail of contingency plans and documented communication. By abstracting a predetermined time frame of care and transcribing into a simple database it is possible to generate a timeline of events with those that contribute to suboptimal care highlighted for discussion at a designated MDT meeting. With minimal training senior medical trainees and midwives or clinical coders can abstract the notes allowing most of the labour intensive work to be done with minimal resource before an MDT meeting. With evolution of electronic health care records (EHCR) it is foreseeable that the burden on data mining will reduce considerably. However, many organisations are a long way off implementing EHCR to this level. Participation in the RCRR process also generates a valuable learning for reviewers. Drawing conclusions on the overall quality of care or avoidability of an outcome is an additional step that requires more implicit reasoning and group consensus. This step can be taken during the MDT meeting if needed but should not distract from reflecting more broadly on lessons amenable to recommendations on ways to optimise care. Hopefully consideration of the MESI RCRR principles outlined above will enable or stimulate obstetric units to undertake more inclusive, frequent and detailed review of major obstetric morbidity. Better organisational learning will most likely be achieved if discussion of RCRR findings has a more reflective focus on ways that care could have been optimised in contrast to debates about avoidable outcomes. References 1.Shah, A et al. Towards optimising local reviews of severe incidents in maternity care: messages from a comparison of local and external reviews. BMJ Qual Saf 2016;0:1-8 2.Carle C, Alexander P, Columb M, Johal J. Design and internal validation of an obstetric early warning score: secondary analysis of the Intensive Care National Audit and Research Centre Case Mix Programme database. Anaesthesia. 2013;68(4):354-67 3.Hutchinson A, et al. A structured judgement method to enhance mortality case note review: development and evaluation. BMJ Qual Saf 2013;22:12 1032-1040. 4.Hogan H, et al. Preventable deaths due to problems in care in English acute hospitals: a retrospective case record review study. BMJ Qual Saf 2012;21:737-745.
Conflict of Interest:
Worthless Statistics in search of Meaningful Conclusion
It appears that these authors believe that variability in the disciplinary rates between states is something that indicates a lack of quality and/or a lack of uniformity of safety measures.
Nothing could be further from the truth.
There are many more reasons affecting a state's disciplinary rates than those controlled for in the study. For just one glaringly obvious example, in certain states and in DC many licensees do not ever set foot in a state, do not touch patients or do anything that could possibly endanger patient safety.
A state MLB is doing its job ONLY when it carefully considers each potential disciplinary case on its own merits and the decisions reached are totally separate and apart from all other cases and decisions in other states. MLB members should never even be exposed to statistics from other states, lest they fall prey to the perennial "Public Citizen" ploy of daring them to play "let's us not be last in the disciplinary contest".
To imply that a narrower spread of disciplinary actions across all states would reflect enhanced patient safety is ludicrous. What that would in FACT suggest is that all state MLBs are "grading on a curve" without regard to actual merit, not bothering to take either their jobs OR their state patients' safety seriously, but simply attending to their statistics and averages. A cynic would say "Oh, it's the 15th of the month, better throw a few more docs under the bus!"
It would also appear that the authors are not aware of the Federation of State Medical Boards, which does everything in its power to promulgate standards and policies for disciplinary activities through its conferences, webinars and publications.
Conflict of Interest:
Why do we love to hate ourselves?
Dhaliwal's comment  on Zwaan et al  nicely refutes what has been called "the hypothesis of special cause"  - the notion that when things turn out wrong, the cognitive processes leading to that outcome must have been fundamentally different (ie, error-prone) from when they turn out right. Dhaliwal's argument recapitulates thinking that is over 100 years old; one of the early contributors to psychology, Ernst Mach, wrote (in 1905): "Knowledge and error flow from the same mental source; only success can tell one from the other" .
What is interesting here is not that the hypothesis of special cause is wrong, but rather the question of why has it been so popular and persistent. What is it about the notion of humans as fundamentally irrational, poor decision-makers that gives this idea such wide appeal? After all, broad acceptance of this sort is not the norm for most psychological or medical research; controversy, argument, or outright disbelief are much more common . Christensen-Szalanski and Beach surveyed decision-making studies in psychology and reported that, although the studies' conclusions were roughly evenly divided between finding good or poor decision-making performance (56% vs 44%), studies reporting human performance as flawed were cited almost 6 times more frequently than those reporting it good. Citations outside of psychology journals were overwhelmingly used to advance the claim that people are poor decision-makers .
One reason for this strange popularity is that the people-are-irrational claim provides benefits for those who have rationality to sell: guideline authors, health care managers, and other proponents of scientific-bureaucratic medicine [6,7]. Another is that it paradoxically provides individual benefits: once we understand the clever puzzles of heuristics and biases problems, even in retrospect, we tend to feel that we must be pretty clever also. And a final, and likely strongest influence, is that it protects organizations and elites: attributing adverse events to flawed mental processes at the front lines serves as a kind of lightning rod, conducting the harmful consequences of bad outcomes down an organizationally safe pathway .
Unfortunately, the history of patient safety to date does not suggest that cautions such as Dhaliwal's will have much effect; such cautions have been raised and ignored before [9-12]. Patient safety's fixation on 'medical error' as the fundament of medical harm serves many (perhaps extraneous) purposes, but is based on an ontological will-of-the-wisp [3,13,14]. Given general agreement on the meagre progress of the patient safety movement to date [15-18], a fundamental re-thinking of our basic premises and hidden assumptions is desperately needed if we are to move forward. And as with many fixations, a sea-change of this sort is not likely to come from within the present patient safety movement, but must come from the outside [19,20]. We can only hope 'these barbarians' challenge us sooner rather than later .
1. Dhaliwal G. Premature closure? Not so fast. BMJ Quality & Safety 2016 bmjqs-2016-005267:online ahead of print.
2. Zwaan L, Monteiro S, Sherbino J, Ilgen J, Howey B, Norman G. Is bias in the eye of the beholder? A vignette study to assess recognition of cognitive biases in clinical case workups. BMJ Quality & Safety 2016.
3. Hollnagel E. Safety-I and Safety-II: The Past and Future of Safety Management. Farnham, UK: Ashgate; 2014, 187 pages.
4. Mach E. Knowledge and Error. Translated by Foulkes P, McCormack TJ. Dordrecht, Netherlands: Reidel Publishing Co; 1905 (English translation 1976), 393 pages.
5. Lopes LL. The Rhetoric of Irrationality. Theory & Psychology 1991;1(1):65-82.
6. Harrison S, Moran M, Wood B. Policy emergence and policy convergence: the case of 'scientific-bureaucratic medicine' in the United States and United Kingdom. The British Journal of Politics & International Relations 2002;4(1):1-24.
7. Wears RL, Hunte GS. Seeing patient safety 'Like a State'. Safety Science 2014;67:50-57.
8. Cook RI, Nemeth C. "Those found responsible have been sacked": some observations on the usefulness of error. Cogn Technol Work 2010;12(1):87-93.
9. Henriksen K, Kaplan H. Hindsight bias, outcome knowledge and adaptive learning. Qual Saf Health Care 2003;12(Suppl 2):ii46-ii50.
10. Dekker SWA. Patient Safety: A Human Factors Approach. Boca Raton, FL: CRC Press; 2011, 250 pages.
11. Hollnagel E. Does human error exist? In: Senders JW, Moray NP, eds. Human Error: Cause, Prediction, and Reduction. Hillsdale, NJ: Lawrence Erlbaum Associates; 1991: pp 153.
12. Wears RL. The error of chasing 'error'. Northeast Florida Medicine 2007;58(3):30-31.
13. Dekker SWA. Is it 1947 yet? http://www.safetydifferently.com/is-it-1947-yet/, accessed 19 May 2015.
14. Woods DD, Dekker SWA, Cook RI, Johannesen L, Sarter N. Behind Human Error. 2nd ed. Farnham, UK: Ashgate; 2010, 271 pages.
15. National Patient Safety Foundation. Free From Harm: Accelerating Patient Safety Improvement Fifteen Years after To Err Is Human. Cambridge, MA: National Patient Safety Foundation; 2015, http://www.npsf.org/custom_form.asp?id=03806127-74DF-40FB-A5F2-238D8BE6C24C, accessed 8 December 2015, 59 pages.
16. Pronovost PJ, Ravitz AD, Stoll RA, Kennedy SB. Transforming Patient Safety: A Sector-Wide Systems Approach: Report of the WISH Patient Safety Forum 2015. Qatar: World Innovation Summit for Health; 2015, http://dpnfts5nbrdps.cloudfront.net/app/media/1430, accessed 18 February 2015, 52 pages.
17. Baker GR, Black G. Beyond the Quick Fix. Toronto, ON: University of Toronto; 2015, http://ihpme.utoronto.ca/wp-content/uploads/2015/11/Beyond-the-Quick-Fix-Baker-2015.pdf, accessed 12 November 2015, 32 pages.
18. Illingworth J. Continuous improvement of patient safety: the case for change in the NHS. London, UK: The Health Foundation; 2015, http://www.health.org.uk/sites/default/files/ContinuousImprovementPatientSafety.pdf, accessed 12 November 2015, 40 pages.
19. De Keyser V, Woods DD. Fixation Errors: Failures to Revise Situation Assessment in Dynamic and Risky Systems. In: Colombo AG, de Bustamante AS, eds. Systems Reliability Assessment: Springer Netherlands; 1990: pp 231-251.
20. Woods DD, Cook RI. Perspectives on human error: hindsight biases and local rationality. In: Durso FT, Nickerson RS, Schvaneveldt RW, et al., eds. Handbook of Applied Cognition. 1st ed. New York, NY: John Wiley & Sons; 1999: pp 141-171.
21. Cavafy C. Waiting for the Barbarians. http://www.cavafy.com/poems/content.asp?id=119&cat=1 . accessed 6 March 2014.
Conflict of Interest:
Statistical Process Control and Interrupted Time Series
I read Fretheim and Tomic's article  with interest as I trained in frequentist stastistics and now work primarily with Stastistical Process Control (SPC) in quality improvement (QI) initiatives.
I concur that there are missed opportunities for using Interrupted Time Series (ITS) in QI; however, I note cautions in doing so:
Regression models applied in ITS often have the assumption of homoscedasticity - i.e. that the process is no more or less variable over time. In evaluation of QI initiatives we are seeking to violate that assumption in that we wish to change a variable and non-performing system into a reliable and highly performing system. This means that the ordinary least squares estimators would not be best linear unbiased, with a risk of overestimating the goodness of fit. This can be addressed using generalised least squares estimators. 
Applying an ITS break point immediately on "intervening" assumes that the intervention is immediately and consistently applied from that point on. This might be true for a total replacement of one drug with another, but is less clear for behavioural change interventions, such as hand hygiene in hospitals. A similar case can be made for care bundle interventions - one may start bundle in all cases, yet all aspects of the bundle may not be consistently conducted.
I suspect that, as with statistics in general, that care is required in selecting one's test, which will depend on the purpose required.
Perhaps a mixed method of SPC and ITS could be preferred?
 A. Fretheim, O.Tomic. Statistical process control and interrupted time series: a golden opportunity for impact evaluation in quality improvement. BMJ Quality and Safety 2015;24:748-752
 A. H. Welsh, R. J. Carroll, D. Ruppert, Fitting Heteroscedastic Regression Models. Journal of the American Statistical Association 1994;89:100-116
Conflict of Interest:
Hand Washing is about Respect for Patients
The paper by Redelmeier and Shafir resonated strongly with me because I have always believed that there are important factors that motivate some physicians to wash their hands while others behave differently. I agree completely that this is a more complex issue than has been previously noted. I always wash my hands in front of patients and have done so for over 40 years. This has very little to do with the risks of healthcare associated infections and much more to do with respect for patients and a commitment to patient-centered care.
I was a pediatric resident at Boston Children's Hospital in the mid- 1970's and as nonchalant about hand washing as everyone else. Fortunately, I had a wonderful mentor whose example changed me. Professor Charles A. Janeway was a renaissance figure in international pediatrics and an excellent teacher. Though in the twilight of his career Prof Janeway would make rounds once a week with one of the senior residents and ward team, examining a patient or two and then discussing the diagnosis and management of each child.
One day we presented the case of a toddler with chronic inflammatory arthritis. As Prof Janeway entered the patient's room, he washed his hands, introducing himself to the child's mother, and then sat and talked calmly with her, once holding her hand when she became tearful. He asked if he might examine her child who was sitting in her lap and clinging to her. Before examining the child he washed his hands a second time. After the exam and further discussions with the mother, he washed his hands again, a third time, before leaving the room.
In a small conference room we began to discuss the case, and one of the students asked Professor Janeway why he washed his hands three times. He slowly looked around at all of us, and said something to the effect of the following.
"I washed my hands on entering and leaving the room because I did not want to bring any infectious agents into or out of the room. I washed my hands again, the second time before examining this frightened child, because there is something about washing hands that sends a message about caring. It is an honor and privilege to practice medicine and hand washing sends a message about respect. The sound of water flowing and the warmth of hands after washing conveys sensitivity and compassion, and patients find this comforting."
I never forgot this lesson and have always washed my hands in front of patients, not so much because of my fears of infection transmission, but because of the message it sends to them. It's not all that complicated really.
Ref: 1. Redelmeier DA, Shafir E. Why even good physicians do not wash their hands. BMJ Qual Saf 2015;24:744-747.
Conflict of Interest:
This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of BMJ
Quality & Safety.
View free sample issue >>
Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.