Table 9

 Factors or conditions facilitating application of SPC*

Examples/explanations
*We found reports of facilitating conditions or factors in 28 articles. The factors formed 23 categories which are grouped into five broad areas.
Information technology and other tools facilitated SPC application
1Easy-to-use tools and resources exist to aid with data management and control chart construction22,30,31,35,45,47,69,70The evolution of information technology has brought many new possibilities for SPC data management, control chart construction and interpretation
2Electronic clinical information systems with SPC capability facilitated data management and SPC application19,25,28,30,34,47,49When designing a system which produced control charts for measures reflecting performance relative to a critical pathway for cardiac surgery, investigators found that a facilitating factor “was the existing electronic medical record, which requires that nurses enter a computerized progress note. ... Integration of the data collection with the generation of a nursing progress note clearly provides the motivation for the bedside nurse to use the system. Moreover, we were able to couple our data collection system with the requirement for an electronic progress note efficiently, so that ... the intranet-based system decreased by 50% the amount of time the nurses spent charting their progress note” (Ratcliffe et al,47 pp 1820–1)
Feedback, training and learning may facilitate SPC application
3User feedback guided changes to facilitate SPC application40“In a focus group session with patients, we learned that many objected to plotting the exact PEFR values over time on a chart. The fear was misplotting – thereby ruining the chart. This insight led to a redesign of the chart in a manner that respected their concern yet did not compromise the chart’s ability to provide the information needed” (Boggs et al,40 p 174)
4Training users in SPC application may be helpful15,26,44,50Reflecting on efforts to apply SPC at a provincial hospital in Indonesia, the author suggests that “(i) the process of data management should be performed by the hospital staff under supervision and with training they should be able to do it without supervision; (ii) a training session for staff on the creation and interpretation of the control chart should be carried out” (Purba,44 p 147)
5Gradual introduction of SPC can allow learning and improvement to support subsequent application and spread25,34,61At a family medicine clinic, the impact of QI including SPC application “had to be demonstrated through initial projects before there was more general support for the process by other residents and faculty within the department” (Ornstein et al,34 p 360)
6Professionals in training led improvement efforts using SPC as part of a clinical scholars programme34“Many of the ideas for patient care improvement projects [using SPC] have evolved from the work and leadership of a group of family residents, pharmacy residents, and faculty that is involved in a unique educational experience known as the clinical scholars program. The program was developed ... as a mechanism to provide the family medicine residents with hands-on opportunities to learn clinical research and quality improvement methods. ... Generally, the project leaders are residents in the clinical scholars program” (Ornstein et al,34 p 351)
7SPC application to clinical topics captured the interest of clinicians34“Clinical topics interest clinicians, and recognition of the potential personal benefits of a project will increase individual commitment and participation” (Ornstein et al,34 p 360)
Patients can have a role in SPC application
8The benefits of SPC application motivated and empowered patients to record and control chart the data35“Adherence in our clinic population is better, both short-term and long-term, with [asthma patient operated control charts of PEFR] than for any PEFR monitoring method we have used in the past. ... It is our assessment that the reasons for high patient adherence with our method include (1) the PEFR measurement per se is simple and can be performed using inexpensive equipment; (2) the charts our patients use are easy to keep and understand; (3) the concepts of signals and not being at-risk are easily understood and appreciated by patients; (4) the chart plays a central role in each patient’s follow-up visit and in telephone calls between visits; (5) because of the latter, patients know we will be monitoring their adherence; (6) patients understand that their chart provides information we need and will use to make decisions that will impact their care; and (7) the quality of the communication with patients is more substantive” (Boggs et al,35 p 561)
9Patients as partners may be better at collecting data for SPC analysis than “human subjects” are in traditional research38“For the control chart analysis used in the current study, the patient has been a partner (the first author of this article). Patients in RCTs usually are seen as ‘human subjects.’ Psychological theory suggests that partners are better data collectors than subjects” (Solodky et al,38 pp AS18–9)
Literature and experts can guide SPC application
10There is literature to guide SPC application15,35“The statistical methods to be used in calculating control charts are well described in the literature.[Ref]” (Re et al,15 p 395)
11Local experts may provide technical assistance for SPC application18,47“At the Indiana University Medical Center several persons from different hospital departments have volunteered to form a cadre of internal statistical process control consultants. The group has come to be known by some as the statistical process control ‘think tank’. No doubt a small group of individuals who are interested in statistics, quality control, and computers can be found in most hospitals. While it is necessary that the large majority of managers and supervisors in any health care organization are knowledgeable about process management and control charts at the concept level of understanding, it takes a relatively few to understand and use such control charts. As internal consultants, these few persons can provide technical assistance when such aid is needed” (McKenzie,18 p 81)
12Collaboration with SPC experts may facilitate application23,34Collaboration between scholars from nursing and experts in industrial measurement error research “has facilitated the conception and completion of a study [of blood pressure measurement error in a nursing clinic] that could not have been accomplished by either discipline alone” (Nelson et al,23 p 27)
Smart application of SPC can be helpful
13Freezing control limits and the centre line at pre-intervention levels helped decipher the impact of interventions59,70“In [a figure], note that we have separated the plot into two periods: before and after the intervention. For calculating the [control limits], one should use only the data from before the intervention. In this fashion, the patient and the clinicians can decipher if the postintervention data exceeds their expectations from historical patterns” (Alemi and Neuhauser,70 p 99)
14Data stratification or disaggregation helped disentangle performance problems and focus improvement efforts37,59“The physicians at Site A needed to develop a theory about why their care was rated lower than at other sites. They theorized that the results might be different for the three service areas: family practice, pediatrics, and ambulatory surgery. Therefore, they developed a control chart to disaggregate the data by service. The results ... for each service were juxtaposed on the same control chart using a technique called rational ordering” (Carey,59 p 80)
15Risk adjustment of SPC data revealed process changes that previously were hidden51,52,71“Before risk adjustment, the process improvement was hidden by the variations in patient risk. Taking the patient risks into account by risk adjustment revealed the significant decrease in [length of stay] during the final 3 months of the 18-month period” (Hart et al,71 p 115)
16Annotations and feedback helped users interpret control charts54,61“For anyone not familiar with the information presented in a control chart, annotations are essential. Annotations will aid in explaining observed variation and explaining changes in the underlying process” (Caron and Neuhauser,54 p 30)
17Keeping control charts up to date increased their utility61“Because the infection control team updates the [control] charts frequently, they can respond more immediately to an increased rate and interact with the ward in a more timely manner” (Curran et al,61 p 15)
18Providing SPC information as locally as possible increased its accuracy and utility26,61“A major reason for the system’s success is that these data charts exist on the PCs at the patient care areas. The information is available to the caregivers and their manager. ... In the [Intermountain Health Care] model, the information is seen first by those who should care the most about its accuracy and reliability. They immediately seek legitimate and operational explanations for anomalies, genuinely showing a high degree of ownership and desire for data integrity” (Shaha,26 p 30)
19Making SPC charts readily available to stakeholders may facilitate SPC application34,65“Feedback to physicians and other providers is an important component of all the ongoing improvement projects. To facilitate this communication, monthly updates to the control charts used to monitor ongoing CQI projects are posted on a large bulletin board [and also distributed at meetings, and sent out by e-mail]” (Ornstein et al,34 p 353)
20Knowing the clinical context helped in the interpretation of control charts25Discussing the issue that blood glucose tests are taken more frequently from patients whose blood glucose level is deranged, investigators argue that they “interpret the charts in context, aware that points are weighted toward derangements” (Oniki et al,25 p 589)
21Evidence of autocorrelation can be detected easily38“Evidence for [autocorrelation—that the preceding observation predicts the next observation] is detected easily by correlating the value of the variable at time t with the value of the variable at time t + 1” (Solodky et al,38 p AS14)
22Concurrent controls can help interpret variation detected using SPC38“Control charts focus on change over time. Adding concurrent control patients can help monitor the effects of unknown exogenous factors that may influence outcomes” (Solodky et al,38 p AS17)
23A key quality indicator helped reflect the functioning of a complex set of processes in a simple way46“Overall process quality cannot be measured directly; we must resort to the use of quality indicators as easy to assess surrogate markers reflecting overall quality. DTNT [door-to-needle-time for patients with acute myocardial infarction who receive a thrombolytic drug] in our example, is a prototype of such an indicator, because it represents the result of a very complex set of processes but nevertheless is simple to measure” (Bonetti et al,46 p 314)