Article Text

PDF

Hospital workers’ perceptions of waste: a qualitative study involving photo-elicitation
  1. Sarah L Goff1,2,3,
  2. Reva Kleppel1,
  3. Peter K Lindenauer1,2,3,
  4. Michael B Rothberg4
  1. 1Department of Medicine, Baystate Medical Center, Springfield, Massachusetts, USA
  2. 2Department of Academic Affairs, Center for Quality of Care Research, Baystate Medical Center, Springfield, Massachusetts, USA
  3. 3Tufts Clinical and Translational Science Institute, Tufts University School of Medicine, Boston, Massachusetts, USA
  4. 4Department of Medicine, Medicine Institute, Cleveland Clinic, Cleveland, Ohio, USA
  1. Correspondence to Dr Sarah L Goff, Department of Medicine, Baystate Medical Center, 280 Chestnut St. Springfield, MA 01199, USA; sarah.goff{at}bhs.org

Abstract

Objectives To elicit sources of waste as viewed by hospital workers.

Design Qualitative study using photo-elicitation, an ethnographic technique for prompting in-depth discussion.

Setting U.S. academic tertiary care hospital.

Participants Physicians, nurses, pharmacists, administrative support personnel, administrators and respiratory therapists.

Methods A purposive sample of personnel at an academic tertiary care hospital was invited to take up to 10 photos of waste. Participants discussed their selections using photos as prompts during in-depth interviews. Transcripts were analysed in an iterative process using grounded theory; open and axial coding was performed, followed by selective and thematic coding to develop major themes and subthemes.

Results Twenty-one participants (nine women, average number of years in field=19.3) took 159 photos. Major themes included types of waste and recommendations to reduce waste. Types of waste comprised four major categories: Time, Materials, Energy and Talent. Participants emphasised time wastage (50% of photos) over other types of waste such as excess utilisation (2.5%). Energy and Talent were novel categories of waste. Recommendations to reduce waste included interventions at the micro-level (eg, individual/ward), meso-level (eg, institution) and macro-level (eg, payor/public policy).

Conclusions The waste hospital workers identified differed from previously described waste both in the types of waste described and the emphasis placed on wasted time. The findings of this study represent a possible need for education of hospital workers about known types of waste, an opportunity to assess the impact of novel types of waste described and an opportunity to intervene to reduce the waste identified.

Statistics from Altmetric.com

Introduction

The USA spends more on healthcare than any other country and has the highest rates of spending growth among developed countries.1 Although economists have argued that increased healthcare spending has economic benefits,2 others have cautioned that it lowers economic growth and employment.3 Additionally, even though increased utilisation accounts for a substantial portion of increased healthcare spending,4 ,5 higher utilisation does not consistently correlate with better patient outcomes.6–8

Efforts to reduce healthcare spending began more than three decades ago.9 ,10 Despite these efforts, a recent Institute of Medicine (IOM) report estimates that there were $765 billion dollars in excess healthcare costs in 2009. These costs were attributed to unnecessary and inefficiently delivered services, excess administrative costs, inflated prices, missed prevention opportunities and fraud.11 Several strategies and tools have been developed in an attempt to reduce these types of waste, including Lean Six Sigma (LSS), which applies manufacturing quality improvement principles,12 the Institute for Healthcare Improvement's (IHI) Hospital Inpatient Waste Identification Toolkit, which provides hospitals with worksheets designed to identify waste in five categories,13 the Agency for Healthcare Research and Quality's (AHRQ) ‘Health Care Efficiency Measures’ and Premier's Waste Dashboard.14 ,15 Although interventions using these strategies have shown promise for improving care processes,16–22 it is less clear how these interventions impact utilisation and patient outcomes.23–25

Photo-elicitation is a common ethnographic technique that involves use of photos to prompt discussions about the area of inquiry during in-depth interviews.26 ,27 Auto-photography (or reflexive photography) refers to a study design in which participants (rather than the investigator) take the photos used in the study.28 ,29 Auto-photography has been used to study questions related to healthcare processes and the impact of building design on patient recovery.30–32

In the study we describe here, we used photo-elicitation with auto-photography to identify sources of waste in the healthcare system from the hospital workers’ perspective. Recommendations for reducing waste were elicited from the healthcare workers as well. Our objective was to further explore how healthcare workers’ perceptions may inform the dialogue about waste reduction in healthcare.

Methods

Study population

Participants were purposively sampled from high volume departments in the hospital (eg, emergency room) and/or those that incur high costs (eg, cardiac catheterisation lab, intensive care units). This sampling strategy was designed to increase the likelihood of eliciting types of waste whose reduction would have substantial impact on costs or patient outcomes. Recommendations from research team members and colleagues were used to decide which individuals from the target departments were to be invited to participate in the study. As a secondary recruitment strategy, participants were asked to refer colleagues (snowball or chain-referral sampling)33 whom they thought would have an interest in participating in the study.

Auto-photography and photo-elicitation

Auto-photography/photo-elicitation was chosen for this study because of the unique way in which still photography engages visual and verbal centres in the brain and stimulates discussion during an in-depth interview.28 ,29 ,34 A photographic image is ‘true’ in the sense that it is a visual representation of what the camera was pointed at.28 ,34 Yet, the photographer's perspective on the picture can be further explored during an in-depth interview.28 ,29 ,35 There is an element of reflexivity, or embedding of the participant's perspective in the research, because the participants choose what to photograph and what to ignore.28 Finally, use of auto-photography also allows the interviewer to more readily join the participant in the discussion since they share an empiric piece of data during the interview.28 ,29

Study design and data collection

Participants were asked to limit themselves to 10 photos representing healthcare waste because repeated sampling of the same subject can lead to ‘intellectually and analytically thin work’.35 Participants took photos with their own digital cameras or a digital study camera. No prompts for what might represent waste were provided so that participants could identify waste as they saw it. Participants were given 2 weeks to take photos. Shortly after the end of a participant's photo-taking period, an in-depth semi-structured interview was held with RK. MR and/or SG were present at the interviews. RK led all interviews using a standardised interview guide (see online supplementary appendix A). Photos were downloaded to a study computer at the time of the interview. During the interview, participants were asked to show each picture they had taken and describe the waste it depicted. The interview guide was updated in an iterative process to probe for emerging themes, such as contributors to waste. Interviews were audio-taped and professionally transcribed. This study was approved by the Baystate Medical Center Institutional Review Board. Informed consent, including permission to keep the photos, was obtained from each participant.

Analysis

Applying grounded theory,36 transcripts of in-depth interviews were reviewed in an iterative process by SG and RK. SG performed open coding on the first four transcripts. Types of waste and recommendations to reduce waste were anticipated as broad themes based upon interview guide questions. RK then coded the same four transcripts independently, using the code book SG had developed. SG and RK then reviewed discrepancies in coding, resolving disagreements through discussion. Analysis proceeded in an iterative process with SG and RK performing open coding on each new transcript and resolving coding discrepancies through discussion. SG and RK performed independent coding on a total of four transcripts during the iterative process to assess inter-rater reliability. Notes (memos) about emerging concepts were made throughout analysis to track developing themes. Secondary coding (axial and selective) was performed by SG during the constant comparative process and secondary codes were refined through discussions with RK and MR. During secondary coding, four categories of waste emerged with associated subcategories. Inductively and deductively derived recommendations for waste also emerged during this stage of coding. Inductively derived recommendations were inferred from comments about postulated contributors to waste while deductively derived recommendations were explicitly stated. After completion of secondary coding, SG performed theoretical coding with RK and MR. During this final stage of analysis, we compared our findings to existing frameworks for healthcare waste to arrive at the final themes. Although theoretical sampling37 was not performed to challenge the theoretical framework, participants were recruited until no new concepts were introduced for four sequential interviews.

Results

Of the 24 individuals invited to participate in this study, three declined. Participants took 159 photos; the number of photos taken ranged from 3 to 13 (median=8). The departments of internal medicine, paediatrics, anesthaesiology, obstetrics and gynaecology, radiology, emergency medicine, healthcare quality, respiratory therapy, pharmacy, and nursing were represented. Degrees held by participants included RN, MD, Bachelor's, PharmD and RRT (respiratory therapy). Forty-three percent were women and the average number of years participants had spent in their field was 19.3 (SD=10.8) with 14.4 (SD=11.7) years at the study institution. Seventy-six percent had some form of administrative experience. Inter-rater agreement reached 85%.

Broad themes identified included types of waste and recommendations to reduce waste. Participants’ recommendations to reduce waste included both deductively and inductively derived elements. Broad themes, categories and subcategories are described in detail below, along with illustrative quotes.

Types of waste

Four major categories of waste were identified: Time, Materials, Energy and Talent, with subcategories identified for Time and Materials. Time was the predominant category with 50% of photos depicting time wastage. (table 1) Most participants (76%) provided examples from several categories of waste, while others provided multiple examples from a single category. Many of the categories and subcategories identified overlapped with the sources of waste described by the IOM (eg, ‘inefficiently delivered services’), LSS (eg, ‘transporting’) and/or the IHI Inpatient Hospital Waste Toolkit (eg, patient ‘flow delay’). (table 1)

Table 1

Categories of waste and their overlap with existing waste frameworks

Categories (bold) and subcategories (italics) are described below. Categories of waste and contributing factors are shown in table 2. Additional pictures with descriptive captions are provided in online supplementary appendix B.

Table 2

Examples of waste and contributing factors with descriptive quotes

Time

A physician (internal medicine, hospitalist) described time as the most ubiquitous source of waste, and shared a picture of a clock when describing this source of waste (figure 1): It's time. It's time. It's our biggest wasted resource … our biggest waste of opportunity.

Figure 1

Time as a wasted resource.

Subcategories of wasted time included time spent searching for needed items (eg, materials, staff), waiting (eg, for slow computers), transporting (eg, materials, patients) or excess processing (eg, duplicate e-mails, repetitive documentation). Participants identified numerous factors that they felt contributed to wasted time, including outdated technology, poorly designed work space, poor communication, false economies, lack of ownership of the problem, and cost centre ‘silos’.

A nurse (internal medicine wards) offered an example of time she wasted searching for patient chairs. She explained that chairs were shared among all medical floors and managers had no incentive to buy new ones or repair broken ones: Fighting about chairs. It's a resource … that we don't have, but it's also the time spent looking [for chairs] … I am wanting to get my patient … out of bed ... and we have no place to put them. And I have no incentive … to buy them [chairs] because I can't keep them [on my unit].

The electronic medical record (EMR) was identified as a source of time wasted waiting by both nurses and physicians (6% of photos). One physician (internal medicine, hospitalist) described the slow processing time of the EMR, while showing photo of the hourglass that appears on the computer screen while it is processing: … The EMR sometimes gets sluggish and slow … I've stared at the hour glass sometimes for as long as a minute … many times a day.

Time wasted transporting patients or materials were identified by several nurses. Following a work re-design, nurses (internal medicine ward) found themselves not only wasting time transporting patients but also searching for wheelchairs to complete this task: So when we had our reduction in workforce [transport services] … their solution was that discharges would … be handled by the nurses … I'll have a wheelchair near my nursing station with a sign on it … “Please do not take, it is for discharge” … People take it.

A number of participants expressed concern that time wasted searching, waiting and transporting resulted in less time for patient care. For example, while discussing time wasted due to duplicate documentation requirements (excess processing), a nurse (internal medicine wards) also expressed concern about the potential impact on patient care. So you are attached [to the computer], it's taking you away from the patient. So you are with the patient less, which we don't like.

Materials

Over utilisation of medical interventions (eg, diagnostic lab tests, procedures and radiological tests) and excess inventory (eg, excess food supplied at meetings) were two subcategories of material waste. Examples of over utilisation comprised 3% of photos. Factors identified as potential contributors to over utilisation included defensive medicine, patient and referring physician expectations, medical education/trainees, and lack of physician knowledge. One physician (cardiologist) expressed strong negative feelings about the over-reliance on stress testing in the evaluation of chest pain: We do entirely too many stress tests plus doctors no longer know how to take a history … It’s easier to order a test than to … take care of a patient, and … use your brain.

A subspecialist in internal medicine felt that overuse of CT scans and MRIs contributes to waste, noting that one patient had seven different head imaging studies done in a relatively short period of time (figure 2).

Figure 2

Excessive use of radiological imaging.

An anesthaesiologist cited the practice of discarding multidose medication vials after a single use because of infection prevention regulations as an example of excess inventory. Another example was a sterile procedure tray being stocked with rarely used items that were then discarded (figure 3).

Figure 3

Single use vials waste medication.

Excess inventory was hypothesised by participants to result from attempts to avoid wasting time searching for items needed during a procedure or from poor planning.

Energy

Eight participants (38%) offered 12 different examples of wasted energy. Some of the examples were similar to potential sources of wasted energy in homes, but on a larger scale; lights and appliances/computers left on when not in use, doors left open during heating and cooling seasons, thermostats set too high in winter and too low in summer, use of old/inefficient appliances and failure to use updated technology such as censored lights. One physician (OB-GYN) commented on energy wasted by lights left on all night (figure 4). … I work on Thursday nights [in a large office building] and when I leave, lights are on, computers are on, the copier stays on. Everything stays on, all night long, every night …

Figure 4

Turning off light switches may reduce wasted energy.

Another physician (oncologist) described amazement at how cold the hospital is kept in the summer and hypothesised that even small changes in thermostat settings could result in substantial savings (figure 5). So I took a picture … of the thermostat … it was, like, 69 or 70 degrees ... I think that we could probably save a fair amount … by having things be a little warmer in the summer and a little cooler in the winter … Could we save something by just making a degree difference? ... I was walking from the hospital through the corridor over to [name of coldest hospital wing] and we could feel this cold breeze coming at you and I went, “Yep, it's the differentials [between the two buildings]” … It's amazing there wasn't a tornado or something.

Figure 5

Adjusting thermostats may reduce wasted energy.

Other sources of wasted energy described by participants were more specific to the healthcare setting: an O2 compressor left on all night while not in use, idling ambulances and use of large shuttle buses at times of day when they are essentially empty. Two participants commented on the use of large, largely empty shuttle buses. One (respiratory therapist) observed: “And then … just the N-Lot bus; the huge buses … sometimes I come in for 11 [AM] and there'll be two people [on] this huge bus.

Lack of accountability and lack of awareness of the potential for savings were cited by participants as potential contributing factors to wasted energy.

Talent

Although less commonly reported than other sources of waste, four participants offered six examples of wasted talent and skill. These examples comprised three broad categories: being required to perform tasks below level of expertise, loss of good employees due to working conditions and failure to hire good candidates because of hiring practices. For example, a pharmacist felt the lack of a phone system that directs the caller to the right person pulled pharmacists away from their responsibilities to field calls that should have been handled by a technician, resulting in delays in patient care: We just have a general phone line, and everyone's first instinct is “let me talk to the important person”. [the pharmacist]. When there’s … four pharmacists… consistently doing orders, and … you constantly take them away, it delays order verification...

While discussing concerns regarding nursing shortages and related additional responsibilities for nurses, a nurse in a management position expressed concern regarding loss of talented nurses because of the working conditions created by the shortages: … I have lost … two of … my best nurses … you have to go home … and know you did the best you could, but when doing the best you could sometimes leaves your nursing care so very wanting, you choose to leave [your job] …, and that's what we're seeing.

Another example of wasted talent was a perceived failure to hire ‘good’ internal trainees who were candidates for faculty positions because of recruitment strategies that were perceived to focus on external candidates. One administrator observed: So … our hospitalist hiring [process] has been less than adequate … there's been a lot of unhappy people [residents] that have gone through and interviewed in … the past couple of years. And … who've not stayed because of their experience [with recruitment].

Participants felt potential contributors to wasted talent included false economies, in which efforts to reduce costs by reducing staff resulted in other sources of waste. Misguided emphasis in hiring practices and poor system design were also considered to be contributors to wasted talent.

Participants’ recommendations to reduce waste

Responsibility for addressing waste lay at three organisational levels: micro-level (eg, individual or ward), meso-level’ (eg, hospital or integrated health system) and macro-level (state or national government and/or insurance companies). Examples of both deductively and inductively derived recommendations made at each of these levels are described below.

Deductively derived recommendations

An example of a micro-level intervention was offered by a physician (OB-GYN) who was disturbed by the amount of energy wasted. She felt that if every person took small steps to conserve energy, the savings would be substantial: … just… please turn the lights off when you're done with the room…

A meso-level recommendation made by a hospitalist was to provide all physicians and nurses in the hospital with cell phones, theorising that text messaging could improve communication.

The few macro-level recommendations explicitly offered were general calls to ‘fix the healthcare system’ with no detailed policy recommendations made.

Inductively derived recommendations

Although not explicitly offered as recommendations, participants’ comments about perceived contributors to waste suggested potential solutions (table 2). These inductively derived recommendations also fit the micro-organisational, meso-organisational, and/or macro-organisational levels.

Micro-level contributors included personal attributes such as ‘inconsideration’, ‘lack of professionalism’, ‘lack of knowledge’, ‘variation in clinical practices’, ‘staff and physician inertia’ and ‘poor communication’.

Meso-level contributors to waste were the most commonly described among the three organisational levels and included elements such as ‘poor inventory management’, ‘poor space management’, ‘costs introduced by trainees’ inexperience’, ‘understaffing’, ‘outdated technology’, ‘poor coordination of administrative responsibilities’, ‘false economies’ and ‘variation in resource needs’.

Macro-level contributors to waste included, ‘policies and regulations’, ‘defensive medicine’, ‘reimbursement structure’ and ‘medical uncertainty’.

Discussion

In this study, we found that healthcare workers’ perceptions of what constitutes waste may differ from the types of waste previously described. Sources of healthcare waste may be broadly categorised as operational or clinical in nature.38 Participants in our study tended to describe operational sources of waste, such as workflow inefficiencies, more than clinical sources of waste such as excess utilisation. Medical errors and fraud, cited by the IOM as major sources of waste, were not described at all. Participants also introduced two novel types of waste that have been less commonly discussed in reference to healthcare waste: energy and talent.

Our participants’ emphasis on operational sources of waste may be explained, in part, by reflexivity, which is inherent to the auto-photography methods used in this study.28 ,29 An example of reflexivity is how the participants’ employment status as healthcare workers may interact with their decisions of what to photograph and what not to photograph.28 ,29 For example, participants may have photographed examples of wasted time more often than problems related to over utilisation because it may be more difficult, either consciously or subconsciously, to call attention to waste you may personally contribute to.28 The use of auto-photography may also have inhibited identification of more abstract forms of clinical waste because they are difficult to capture with still photography. One further potential explanation for the emphasis on organisational waste seen in our study was that some sources of waste may not have yet entered hospital workers’ consciousness because they are not readily apparent (eg, price of services out of proportion to value). If the latter is true, this may support prior work calling for physicians to learn about waste in healthcare and to participate more fully in waste reduction efforts.4

The novel types of healthcare waste introduced in this study warrant further discussion. ‘Wasted talent’ has not been commonly described in prior work on healthcare waste. However, this source of waste was described in three contexts in our study: workers required to perform duties below their skill level, loss of ‘good’ employees due to a perception of a suboptimal work environment and failure to retain trainees because of ineffective hiring practices. Asking employees to perform duties below their skill level on a regular basis introduces inefficiencies and potentially job dissatisfaction39 This may then lead to disengagement and reduced productivity,39 ,40 which may then have a negative impact on patient care and patient satisfaction.40 With healthcare reimbursement increasingly tied in part to patient satisfaction,41 this may have important financial ramifications. Employee turnover generates costs for recruitment and training of new employees,42 and may impact the morale or remaining employees,43 leading to disengagement. Further economic assessment of costs related to ‘wasted talent’ may demonstrate a potential for cost-savings. A second novel type of waste discussed was ‘wasted energy’. Even though energy conservation has long been in the public consciousness, it is not commonly cited as a type of waste to be addressed in healthcare. Several participants in our study described wasted energy, such as lights left on and inappropriate thermostat settings, as a potential area for cost-savings for the healthcare system. Although no large-scale efforts to reduce energy consumption in healthcare have been described, Practice Green Health is an organisation devoted to reducing various sources of waste, including energy, in the healthcare industry.44 Additionally, a recent study showed that working in a building specifically designed to reduce energy consumption led to greater conservation behaviours by people working in the building.45

The majority of participants’ recommendations to reduce waste required intervention at the micro- organisational or meso-organisational level, which may be in part due to the enormity of macro-level issues in U.S. healthcare. Recommendations to reduce waste that may be most amenable to intervention included improving planning, reducing processing times on the current EMR, and organising physical space to be more efficient. Contributors to waste that may be more challenging to address include workers’ attitudes, such as inertia, inconsideration and lack of ownership. Use of photo-elicitation may strengthen future interventions aimed at reducing waste, particularly when used in conjunction with methods such as process mapping, because this method increases participant engagement and allows for a deeper understanding of their perspective.34

A number of recommendations to reduce the waste identified appeared feasible and relatively straightforward (eg, as asking people to turn off lights and using text messages for communication between doctors). However, ‘fixing’ many of the waste issues described in this study would be a complex undertaking, requiring re-engineering systems that have been in place for years and that often involve multiple stakeholders. Moreover, waste reduction efforts may generate other forms of waste. For example, if all potentially necessary items for a procedure tray were not prepackaged as recommended by one study participant, time would be wasted searching for these items when they are needed. Although EMRs have demonstrated benefits,46 ,47 improving the EMR as suggested by participants in this study, would likely consume already strained information services resources. Replacing the EMR would require institutional dollars and substantial time investment by users to learn the new system. These findings highlight the importance of identifying potential challenges and proceeding with caution when intervening to reduce waste, precautions that have been previously advised.48

This study has limitations. First, auto-photography/photo-elicitation provided a unique method for identifying healthcare waste but may have restricted examples of waste identified to those items that were easily photographed, represented waste that was most frequently encountered or that were most personally intrusive. Second, this was a single centre study, and examples of waste provided at other academic centres or community hospitals may differ. Third, although our purposive sample came from a broad range of departments with varying job descriptions and we conducted interviews until no new concepts emerged, additional participants may have contributed novel concepts.

Despite decades of efforts to improve the value of healthcare in the U.S., systemic waste remains a daunting challenge. The feasibility and merit of the interventions recommended by participants in this study should be explored further to see if they may meaningfully impact waste, particularly at the micro-level and meso-level. This study also demonstrates the potential for using photo-elicitation to study other areas of healthcare, such as to explore patient satisfaction, or to illuminate the challenges that patients in special populations may encounter during interface with the healthcare system.

View Abstract

Footnotes

  • This work was presented in poster form at the Pediatric Academic Society Meeting, Boston, MA, April 29, 2012, as an oral presentation at the Society for General Internal Medicine Meeting, Orlando, FL, May 11, 2012 and will be presented in poster form at the AcademyHealth Research Meeting June 2013.

  • Contributors SLG MBR and RK contributed to the study concept and design, data collection, analysis, and interpretation and manuscript preparation and review. PKL contributed to data interpretation and manuscript preparation and review.

  • Funding This research has been funded by the National Center for Research Resources (KL-2 RR025751) and the National Center for Advancing Translational Sciences (KL-2 TR000074), National Institutes of Health.

  • Competing interests None.

  • Ethics approval Baystate Medical Center IRB.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

References