Lean thinking across a hospital: redesigning care at the: Flinders Medical Centre.

By: Be n-Tovim, David I.,Bassham, Jane E.,Bolch, D enise,Martin, Mar garet A.,D ougherty , Melissa ,Szwarcbord, Michael Publica tion: Australian Health Rev iew Date: Thursday, February 1 2007

Abstra ct Lean thinking is a method for organising comple x produ ction processes so as to encourage flow and reduce waste. W hile the prin ciples of lean thinking were developed in the manufacturing sector, there is in creasing interest in its application in hea lth ca re. This case history do cuments the introd uction and development of Redesign ing C are, a lean thinking -based p rogra m to redesign ca re processes across a tea ching gene ral hospital. Redesigning Care has produ ced substantial benefits o ver the first two-and-a-half years of its imp lementation, making care both safe r and more a ccessible . Redesign ing C are has not been aimed at chang ing the spe cifics of clin ica l practice . Rathe r, it has been con cerned with imp roving the flow of patients throug h clin ical and other systems . C oncepts that emerged in the manufacturing sector have been readily translatab le into health care. Lean thinking ma y pla y an important role in the reform of health care in Austra lia and e lsewhere. Aust Health Re v 2007: 31(1): 10-15 ********** THIS PAPER DESCRIBE S the introdu ction and early results of the Redesign ing C are Program at the Flin ders Medical Centre. Redesign ing C are explicitly app lies lean th inking (1) to health ca re. Lean thinking is a codification of man ufacturing te chn iques pioneered b y the To yota Motor Compan y. Du ring a la rge scale

analysis of motor manu fa cture rs, (2) the p rodu ctio n processes developed b y the To yota Motor Compan y were identified as being so different from those of other large car-ma kers as to constitute a new manufacturing methodo logy, re ified b y the authors as lean thin king.

Lean thinking and redesigning care Lean thinking is an atte mpt to correct the delays and wasteful reduplications that chara cterise man y mass production processes. The issues have rema ined essentia lly un changed sin ce Henry Ford developed the mass production methodology in the 1920's. In orde r to produce large nu mbers of differing kinds o f vehicles, Fo rd arranged his fabricatio n machinery into production villages dedicated to specific fu nctions. The inevitable hiccups in p roduction within and between production villages were managed b y mainta in ing large bu ffe rs of parts at every stage in prod uction. Variations in customer demand were managed by creating banks of finished goods that were forced on dealers (and w hich, when unsold , w ere moved by means of special p romotions ). The costs o f holding large amoun ts of un finished goods w ere contained by reducing the cost per production step b y any means possible , the buffer sto cks managing any resultant dela ys in produ ction . Nowadays, this may mean prod ucing components in widely separated lo cations . Lean thinking facilitates moving from mass to flow production. Flow produ ction aims to d ramatically decrease the time taken to p roduce goods by arranging the re levant machinery in p rocess sequen ces, rapidly changing a machine's w orking p rocess to match changing process requirements so that goods can be made in response to customer de mand rather than to a p rearranged prod uction schedule . C osts are conta ined b y min imising buffer stocks, by rapidly identifying and avoiding w asteful e rro rs in produ ction, and by a voiding wasteful over-prod uction and over-processing. These

latter steps also lead to dramatic impro vements in the quality o f the goods produ ced. The basic p rincip les unde rlying the imp lementation o f lean thinking are laid out in the Box. Interest is grow ing in the potential utility of lean th in king in health care, (3,4) where the need to improve the flow of patients throug h hospitals and health services is beco ming increasingly urgent. Redesig ning Care is using lean thin king to imp rove flow and redu ce waste in co re clin ica l and support services across a whole hospital. It is not, however, con cerned w ith attemptin g to influence the pro fessional co ntent o f clinical encounte rs. That is deemed to be outside the scope o f the program, which is primarily concerned with flow and logistics. This case history may be of some interest to othe rs seeking to transform their hospital or health service using similar methods. The key principles of lean thinking

* Specify the value desired b y the customer * Identify the value stream for each produ ct or service providing that value, and challenge all the wasted steps * Make the produ ct or service flow continuously * Introdu ce pull between all the steps where continuous flow is impossible * Manage towards perfection so that the n umbe r o f steps and the amount of time and in formation transfer needed to serve the custo mer continually fall

S etting The Flinders Medica l Centre is a 500-bed teaching general hosp ital in the southern suburbs of Adelaide. Flinders is a " cradle-to-gra ve" institution, pro viding a co mplete range of secondary and tertiary services to a population o f a round 300 000. It is the largest member of a de-facto consortiu m o f hospital and co mmunity health service providers that also includes a sma lle r general hospita l and a commun ity hospital. The Flinders Medical Centre is the prima ry regional provider of time urgent, comp lex ca re of all kinds . The Emergen cy Department is busy, seeing some 50 000 patients per year, o f whom around 40% require hospital admissio n. S equence of events By mid 2003, the Flinders Medical Centre Emergen cy Department had beco me so congested that patients were regu larly o verflowing into the nearb y re cove ry area o f the operating theatre suite, disrupting the work of both the Eme rgency Department an d the Division of Surgery. Cancellations of ele ctive work were pervasive, surgical training schemes were under scrutin y, the sa fety of care in the Emergency Department was becoming compromised, and high levels of staff turnover were undermining the viability of key clin ical services. These difficu lties had not arisen suddenly; nor were they a consequence of unusual leve ls of de mand. Flinders w as struggling to fulfil the predictable de mands of the popu lation served. The clin ical staff at the Flin ders Medical Centre a re energetic and well motivated and had adopted standard p ractices to dimin ish congestion , (5) but without sustained bene fit. W hat w as needed was to do something that the staff d id no t yet know how to do. The then-hospital board agreed to p rovide non-operatio nal fu nds to support a program o f hospital redesign , the e xact nature o f w hich was yet to be clarified.

Redesigning emergency department flows Two of us (DBT and MD) had started working w ith the E mergen cy Department staff analysing why safe and sustainable care was so hard to provide. Making little headway, we came across a description of process mapping on a National Health Se rvice (NHS) Modernisation Agen cy website. (6) In lean th inkin g te rms, the process is the end-to-end sequen ce of steps required to transfo rm a raw material to a fin ished product (1) and process mapping is the name given to the creation o f an end -to -end flow diagram of the steps involved. Taking the patient's symptoms (1) at the point of presentation as the "raw material" and the patient's journe y from arriva l th rough to exit fro m the Emergency Department as the "produ ct", w e resolved to map the steps involved in the patient journe y throug h the Emergency Department. W e gathered a large multid isciplina ry group of Emergency Department sta ff and started to work our way through the journeys of patients who were either discha rged directly fro m the Emergency Department, or w ho needed admitting to hospital. The care processes involved were described as the staff saw them. Seve ral sessions w ere needed to do cument the steps invo lved in the patien t journe y through the department. The mapping sessions had a profound impact on a ll invo lved. They created a shared aw areness of how chaotic the care pro cesses had become, and generated support to change p rocesses within the Emergen cy Department irrespective of what was being done elsewhere in the hospital. How exactly to do th is w as still not clear, and the search fo r an imp rovement model began in earnest. A small group of sen ior staff made a brief visit to London, hosted b y the NHS Modern isation Agency (sin ce dissolved). The itinera ry included visits to a numbe r o f hosp itals, and d iscussions with Modernisation Agency staff, one of whom also spent severa l days in Adelaide advising the hosp ital on the structure of an improvement program.

The United Kingdom visit demonstrated that real changes cou ld be made to the o rganisation of care within emergency departments, and that those changes could have a profound impact on congestion within those depa rtments . Modern isation Agency sta ff also exposed the Flinders group to the con cepts of lean th inking. Follow ing the trip to the U K, the Director of the Emergency Department proposed a radical restructuring of the w ay patients flow ed through the Emergen cy Department at the Flinders Medical C entre. The mapp ing had de monstrated that attempting to p rio ritise care by means of the Australasian Triage Scale , a five point measure of patient a cu ity, materially contributed to the co mp lexity of patient a llocations w ithin the department. The sta ff w ere continually attempting to respond to the d istress o f patients who were "bumped" out of order fro m the ir place in the notiona l q ueue when a patient w ho arrived after them was seen before them because the y were in a d ifferent triage category. Ad ho c and ha rd to manage strategies were being used to try to push patients th rough when the build -up o f " bumped" patients became e xcessive . The new flows involved brea kin g away from using the triage s core as a method fo r prioritising care w ithin the department. Instead, patients would be assessed by a triage nu rse w ho, w hile a llocating a triage score , w ould also indicate whether in h is or her judg ment the patient w as likely to be ad mitted to hospital or to return home dire ctly from the department. Each stream of patients (likely to be discha rged, likely to be ad mitted) was to be aligned with a separate team of nurses and docto rs in specific areas of the department. In the absence of a threat to life and limb , patients w ere to be seen in order of a rrival. Initially, this w as only if they w ere likely to go home, but subsequently, the p roposal was w idened to include a ll adult patients. Staff received brief orientation to "strea ming", as the new processes came to be described, and it was initiated towards the end of November 2003. The impact was immediate . At the end o f the first

day, there was a disce rnib le lessening of the chaos w ithin the department, and this sense of increased control has continued. Streaming has been well suppo rted b y the staff and has been mainta ined continuo usly since its introd uction. A clear indication of the in creased acceptability of the ca re provided was the immed iate halving of the nu mbers o f patients leaving the department without completing their care . "D id -not-waits" as a percentage of arrivals fell from 7% o f a ll arriva ls to just over 3% and have been mainta ined at that le vel. Streaming also de creased congestion by de creasing the ove rall time patients spent in the department. The average time that patients spend in the department was reduced by 48 minu tes in the first year after imp lementation (bring ing the a verage time spent in the depa rtmen t from 5.7 hours down to 5 hours). The next year saw a 10% increase in the nu mbers of patients attend ing the department, but the decrease in average time in the department was not only mainta ined, it was further redu ced by 6 minu tes. Lean thinking across the hospital The concepts behind streaming de rive d ire ctly from lean th in king. The redesign began with the identification of "patient-care fa milies". Patient-care fa milies are g roups o f patients w hose care pro cesses overall a re sufficiently simila r to each other, yet different from those required by other patient-care families , to be managed together. In this case, the patient-care families w ere "like ly to go ho me", and "likely to be ad mitted to hospital". The sum of the steps needed to complete the jou rne y o f each patient-care family is known, in lean thin king terms , as the value stream. Mapping care processes fro m beginn ing to end allow ed us to "see" patient ca re families and their value streams and to identify w asteful delays and reduplication along the journey. Lean thinking is focused on imp ro ving flow by simplifying produ ction processes, lining up the steps in a value stream so that a steady production rhythm can be achieved. In the Emergen cy Department, reduction o f w aste and improvement in

flow w as achie ved b y creatin g produ ctio n "cells" aligned w ith value streams. Ea ch cell fo cused on a particu lar patient-ca re family and completed work as it arose rather than queu ing patients and then treating the m in batches. The early su ccess of th is intervention w as sufficient to con firm the value of testing the application of lean thinking to co re clin ica l and support services throughout the hospital. The p rogra m was called Redesign ing C are, and the small team o f a part-time director and three fulltime clinical facilitators (all senio r nurses ), suppo rted b y the senior managers in the hospital, set about increasing their knowledge of lean thin king and de velop ing a structured approach to the implementation of lean thin king across the hospital. Initially, Redesign ing C are prog rams were aligned with three broad streams of work (emergency, med ical and su rgica l), each headed b y a senior clinician and each w ith a sponsor from the senior hospital executive. O ver time, the range e xpanded to in clu de support services, mental health , and transition to community ca re. Specific progra ms of w ork a re s coped as to the beginning and end of the patient (o r other pro cess) journeys involved, and a s coping document is ag reed to by the key stakeholders . Process flow mapping and tracking of real-life patien t journe ys are then used to create a detailed picture of how the work is done now (the cu rrent state) and to generate acceptance o f the need for change. A series of "plan-do-study-a ct" cycles a re then initiated based on the imp rovement opportunities that "fall out" of the mapping pro cess. The cycles a re de veloped an d u ndertaken by work groups of staff invo lved, with facilitator assistance. Ta rgets are deve loped by the groups and are mon itored continuously. Formal evaluations at designated points set the scene fo r the most difficult challenge of all: ma king change susta inable in the long run--making the new way the "way we do it rou nd here". Lean thinking con cepts en courage health care p roviders to think about the patient journey fro m arrival to d ischa rge as a co mp lete

care process rather than as a series of dis connected steps. As staff in hospitals and health services, w e tend to be "po int optimisers", focusing on do ing the work in fron t o f us as best w e can , igno ring the impa ct that changes to a step may have on the steps on either side. A clear e xamp le was provided early on in our redesign a ctivities. W e were mapping ou t the mo vement of patients through a large inpatient service when it became clear that patients treated within this service co mmonly spend at least half a day longe r in hospital than necessary because they could not be discharged w ithout a date for a crucial fo llow-up test in a hospital clinical laboratory. That laboratory was under such p ressure to perform tests that it had put the fu nding for its re ceptionist against a new laboratory technician. The net result was that appointments cou ld o nly be made when a laboratory staff member was free to pick u p messages left on an answering s ystem. Getting appoin tments was very difficult, leading to delays in discharg ing patients, which in tu rn increased congestio n in the Emergen cy Depa rtment while newly arrived patients w aited for a bed . Redesigning Care across the hospital Fro m its inception , Redesign ing C are was seen as a change progra m. Support for the program has been built by communicating the methodology and the results in many different ways. Important elements ha ve been "lean thinking" days in which the basic con cepts ha ve been introduced to la rge numbe rs of staff. The re is also a more intensive program of exposure to lean th inking of staff from designated areas who will be key participants in specific progra ms of w ork. By now, hundreds of staff across the hosp ital are invo lved in redesign activities o f one fo rm or another. A hospita l is su ch a diverse entity that it may be hard to know where to begin a program of redesign. However, the pressures generated by the emergency ca re of patients w ere such that they

had to be attended to. But a hospita l is a dynamic entity. One patient cannot be ad mitted u nless a previous patient has been discha rged. Initial mapping w ithin the med ica l an d surgical strea ms indicated that care processes could commonly be separated out into those required by patients who would spend relatively short periods (up to 72 hou rs ) in the hospital, and those required by longer staying patients. A sho rt-stay medical-surgical ward of so me 20 beds was developed for the majo rity of patien ts admitted as an e mergency and pred icted to spend a short time in hospital. This unit now accommodates around one in fou r patients admitted to the hospita l. The staff in the medical and su rgical wards no longer ha ve to split their attention between the comp lex ca re needs o f longer staying patients and the administrative and organisationa l tasks in volved in moving patients rap idly th rough the hospital. The capacity ga ined by this de velop ment enab led the hospital's sma ll elective surgery progra m to return to fu ll fun ctioning . Surg ical train ing s che mes ceased to be under threat, building further support fo r the program. Importantly, the rate of serious adverse even ts repo rted to the hospital insu rers has halved sin ce the Redesign ing Care prog ram began, and the w idespread take-up of clinical imp rovement progra ms a cross key clinical divis ions has also been an important contributor to enhan ced safety across the hospital. Other important Redesign ing C are in itiatives ha ve included : redesigning the p rovision of medication at discharge, ha lving the time taken to p ro vide that med ication; substan tial changes to bed management pro cesses; and redesigning the flow of longer staying medical patients. The latter program has reduced the average length of stay in the la rge genera l medical service by around 1 day of stay. While it has not been the primary fo cus of the Redesign ing C are progra m, in the current financial year the hospital is providing care sufficiently cost effe ctively to be able to dire ct modest savings from

its operational budget into enhanced equip ment replace ment and staffing. Th is is the first time in many years that this has been possible. Problems, conflicts and constraints The Redesigning Care Program is a major change program and as such w ill ine vitab ly come up against a wide variety o f difficulties. One of the mo re thoug ht pro voking of these has been the cha llenge offered to e xisting middle and senior managers. In gene ral, hea lth care managers are chosen fo r the ir p roblem solving s kills. The most successful excel at " fire -fighting" and enjoy the drama involved. But a basic maxim of lean thinking is not to start with a solution, but to go to the workpla ce, unde rstand how the work is done and look for root causes of delays and other impediments to flow. There is therefore a tension between the somew hat painstaking, bottom-up approach emp loyed by Redesig ning Care, and the mo re usua l "command and control" p rocess adopted by health care managers who, once a prob lem has been identified, see their ro le as co ming up with a so lution that front-line staff then have to imp lement. Early closure a nd starting with a solution is not confined to hospital managers. It is pervasive at every le vel in the health syste m. Lean thin king requires managers to ensure that a de cis ion gets made , rather than ma ke every decisio n. Actin g as a fa cilitator to de cis ion making is not easy, and the te mptation to regress to knee-jerk proble m solving seems ever present. Discussion Patients are not cars , and providing good clin ical care invo lves compassion and empathy as well as cognitive and organisatio nal skills. Acknowledging that, it is still possib le to con ceptualise patient journe ys as lengthy sequences of specific trans formative steps strung along de-facto p rodu ction lines spread throug hout hospitals and health services. Lean thinking is not abo ut influencing the content of those moments w hen patients and staff are in contact. It

is about giving more time fo r those mo ments, ma king them easier to perform and less prone to error, by simplifying sequences , making w hat has to be do ne more transparent, re moving reduplicative and unne cessary steps, and making hard-to -pe rfo rm steps easier to get righ t. At an operational level, health care p ro cesses are almost never designed end-to-end . The y e volve s low ly, each component e volvin g within its own niche or process village, and w ithout necessarily taking acco unt of the impact on steps up and down the line. The Flinders Medical C entre has been using lean thinking to ma ke a start on design ing w hole sequences of care--not s imply to provide the care that is rig ht, bu t right first time, fo r the right patient, at the righ t p lace, an d at the right time. As we do so, the s ize of the challenge and the potential benefits of success be come clear. Redesign ing C are has made a start, and it is only a start, on this important task.

Com peting interests The authors declare that they have no com peting interests. (Received 3/07/06, revised 4/09/06, accepted 1/10/06) References (1) Womack J, Jones D. Lean thinking. Banish w aste and create wealth in your corporation. London: Simon and Schuster, 1996. (2) Womack J, Jones D, Roos D. The machine that changed the world. New York: Rawson Associates, 1990. (3) Reinertsen JL. Intervie w with Gary Kaplan. Qual Saf Health Care 2006; 15: 156-8. (4) Jones D, M itchell A. Lean thinking for the NHS. London: NH S Confederation, 2006. (5) Bartlett J, Cameron P, Cisera M. The Victorian em ergency department collaboration. Int J Q ual Health Care 2002; 14: 463-70. (6) National and Primary Care Trust Developm ent Programme. NHS M odernisation Agency Dem and Management Group. The Big W izard. Available at: www.natpact.nhs.uk/demand_managem ent/wizards/big_wizard/downloads.php (accessed Jun 2006).

What is k nown about the topic? Lean thinking, while developed in the manufacturing sector, appears to be a relevant technique for redesigning hospital care. What does this paper add? This paper provides a case study of the implementation of lean thinking, initially in the Emergency Departm ent, and then throughout Flinders M edical Centre. What are the implications for practitioners? Using the principles of lean thinking, practitioners are encouraged to explore the value that process components provide to patients. Using a participative approach, the processes can then be streamlined to improve patient flow. David I Ben-Tovim, PhD, M B BS, MRCPsych, FRACNZCP, Director, Clinical Epidem iology and Redesigning Care Units Jane E Bassham, RN, BN, Clinical Facilitator, Redesigning Care Denise Bolch, RN, RM, BN, MBA, Deputy Di rector, Redesigning Care M argaret A M artin, RN, DipAppSc( Nu rsing), BN, M HSS, Deputy Director, Division of Surgery and Specialty Services M elissa Dougherty, BA(Hons) Psychology, Clinical Facilitato r, Redesigning Care M ichael Szw rcbo rd, BSc, BSocAdmin, FAIM, Executive Director, Acu te Servi ces, a Southern Adelaide Health Services; and General Manager Flinders Medical Centre, Adelaide, SA. Correspondence: Professor David I Ben-Tovim , Flinders Medical Centre, Bedford Park, Adelaide, SA 5042. david.ben-tovim@fm c.sa.gov.au