Figure - available from: BMC Medical Research Methodology
This content is subject to copyright. Terms and conditions apply.
Example control charts. a random variation. b Non-random variation caused by a large, possibly transient, shift in data identified by one data point being outside the upper control limit. c Non-random variation caused by a sustained moderate shift in data identified by an unusually long run of 13 data points below the centre line (Western Electric rule 4 and Anhoej rule 1) and unusually few crossing (Anhoej rule 2)
Source publication
Background:
The aim of this study was to quantify and compare the diagnostic value of The Western Electric (WE) statistical process control (SPC) chart rules and the Anhoej rules for detection of non-random variation in time series data in order to make recommendations for their application in practice.
Methods:
SPC charts are point-and-line gra...
Similar publications
p> Purpose: The purpose of this paper is to present preliminary research in statistical process control (SPC) of short run and small mixed batches (SR-SMB) at the organization producing bakery equipment.
Methodology/Approach: The starting point of the research is a literary survey of possibilities of using SPC for SR-SMB and analysis of the curren...
Citations
... SPC is a methodology for monitoring and controlling processes to ensure that they operate efficiently and produce products or services of consistent quality. It involves collecting and analyzing data from the process in order to understand its variability and make informed decisions about process adjustments and improvements [21]. ...
The production of high-quality products and efficient manufacturing processes in modern environments, where processes vary widely, is one of the most crucial issues today. Statistical process control (SPC) and process mining (PM) effectively trace and enhance the manufacturing processes. In this direction, this paper proposes an innovative approach involving SPC and PM strategies to empower the manufacturing environment. SPC monitors key performance indicators (KPIs) and identifies out-of-control processes that deviate from specification limits, while PM discovery techniques are applied for those abnormal processes to extract the actual process flow from event logs and model it using Petri nets. Different enhancement techniques in PM, such as decision rules and root cause analysis, are then used to return the process to control and prevent future deviations. The application of the integrated SPC–PM approach is shown through case studies of production processes. SPC charts found that over 6% of processes exceeded specification limits. At the same time, PM methodologies revealed that prolonged times for the ‘Quality Control’ activity is the fundamental factor increasing the cycle time. Moreover, decision tree analysis provides rules for decreasing the cycle times of unbalanced processes. The absence of a transition from the ‘Return from Waiting’ activity to ‘Packing and Shipment’ is a critical factor in decreasing cycle times, as is the shift information. Our newly proposed methodology, which combines process analysis from PM with statistical monitoring from SPC, ensures operational excellence and consistent quality in manufacturing. This study illustrates the application of the proposed methodology through a case study in production processes, highlighting its effectiveness in identifying and addressing process deviations.
... Research has shown that using an XmR chart with this set of rules effectively identifies statistically significant outliers and trends. 6 ...
Objective:
This study aimed to examine the patterns of complaints filed against physicians in Rhode Island, investigate the factors associated with complaint rates and outcomes, and assess the impact of the implementation of a new Framework for Just Culture.
Methods:
Complaint data from the Rhode Island Department of Health's complaint tracker and physician licensing database were analyzed for the period of 2018 to 2020. Descriptive and statistical process control analyses were conducted to assess complaint rates, investigation rates, and adverse outcomes.
Results:
Over the three-year period, 1672 complaints were filed against Rhode Island physicians, with approximately 40% of complaints being opened for investigation. The implementation of the Framework for Just Culture coincided with a sustained decrease in the rate of complaints opened. Failure to meet the minimum standard of care was the most common allegation, and male physicians and those aged 40-50 were more likely to have complaints filed against them.
Conclusions:
The study highlights the importance of complaint investigations in upholding standards for medical licensure and clinical competence. The Framework for Just Culture may have influenced the investigation process, resulting in fewer investigations opened without compromising the identification of cases requiring disciplinary action. These findings provide insights into physician accountability and the need for ongoing monitoring and improvement in complaint handling systems.
... We analyzed the SPC charts using a combination of visual inspection and three statistical tests (rules) for non-random variation. These include tests for unusual long runs of data points above or below the center line (rule 1), unusually few crossings of the center line (rule 2), and data points outside the control limits (rule 3) [18]. ...
Purpose
Placenta-mediated pregnancy complications, like growth restriction and hypertensive disorders, are leading causes of maternal, fetal and neonatal morbidity and mortality in high-income countries. The purpose was to investigate if there is a seasonal variation in placenta-mediated pregnancy complications (small for gestational age, intrauterine growth restriction, preeclampsia, preterm birth and intrauterine fetal death).
Methods
This is a Danish cohort study including all singleton deliveries at gestational week 22 up to and including week 41 conceived from December 2006 to November 2016 (N = 555,459). We used statistical process control charts to visualize data and to test for patterns of non-random variation in data over time for pregnancies with risk factors (BMI, diabetes, in vitro fertilization, maternal age > 40 years, primipara, previous caesarean and smoking) and each of the following outcome: fetal growth restriction, hypertensive disorders, preterm birth and intrauterine fetal death. The study was approved by the Danish Data Protection agency; REG-039-2019.
Results
We found a seasonal pattern in hypertensive disorders during pregnancy with dips in pregnancies conceived in the fall season and highest risk by conception in the spring and summer season. We found no apparent seasonality in cases of preterm delivery, small for gestational age and intrauterine mortality. Individual risk factors (e.g. smoking and obesity) for placenta-mediated complicated over time were in consistency with the general trends.
Conclusions
We found a significant seasonal variation in the risk of hypertensive disorders of pregnancy with highest risk by conception in the spring and summer season. This study found no seasonal variation in other placenta-mediated complications.
... Using an XmR chart with this set of rules has been shown to perform well at identifying both statistically significant outliers as well as trends. 12 Rates that fall outside these control limits and rules would suggest a special cause variation, meaning a change in the process of malpractice. Actual verdicts were rare, at <5, because of RIDOH's small numbers policy, the specific number is not reported. ...
Objectives:
To determine the rates and characteristics of physicians with medical malpractice adverse outcomes in Rhode Island.
Methods:
A descriptive epidemiologic study of medical malpractice claims from 2008-2018 aggregated by the Board of Medical Licensure and Discipline of the Rhode Island Department of Health. To examine the demographic characteristics of physician malpractice cases we reviewed 10 years of data from Rhode Island medical malpractice lawsuits that were resolved, in whole or in part, via payment to the plaintiff.
Results:
Over this 10-year period, there were 460 such cases, 88% of which involved a male physician and 48% of which involved surgical category specialists. Few cases, 17.6% of payments, were over one million dollars, and the mean payment value across all cases was $517,104. The rate of paid claims was found to be stable over the period studied.
... Data within this line are acceptable, whereas data between ± 2SD and ± 3SD are considered borderline and no more than 1 data-point is allowed within this limit. Values that fall above ± 3SD are considered not acceptable [21]. The tested compounds are cellobiose, glucose, xylose, arabinose, acetic acid, formic acid, lactic acid and levulinic acid, and all data falls within ± 2SD, though none falls beyond the ± 2SD. ...
Oleaginous yeast, cultured on second-generation lignocellulosic resources, has the potential to be a key part of the future energy sector. However, the multiple unit operations necessary to produce concentrated hydrolysates, with a minimum of fermentation inhibitors, limit the applicability to date. In this study, a simple microwave-assisted hydrothermal pre-treatment step of oak or beech sawdust was deployed to produce an oligosaccharide-rich hydrolysate. This was then catabolised by the oleaginous yeast, Metschnikowia pulcherrima, avoiding the need for costly enzymatic or further chemical steps in the processing. Up to 85% of the sawdust’s hemicelluloses could be solubilised under these conditions, and 8 g/L DCW yeast with a 42% lipid content produced. While a number of studies have demonstrated that oleaginous yeasts possess high inhibitor tolerance, using this real lignocellulosic hydrolysate, we demonstrate that lipid production is actually very sensitive to inhibitor and carbon availability, and the optimal system is not the one that gives the highest hydrolysate or cell biomass. Indeed, the yeast was shown to detoxify the inhibitors in the process, but at high inhibitor loading, this leads to very poor lipid production, especially at high furfural levels. These findings clearly highlight the importance of considering multiple variables when real, complex lignocellulosic media are involved, tuning process conditions based on the desired fermentation outcomes.
... (We use the NHS SPC tools, 26 which have a limit to how many datapoints can be entered; plus the SPC pattern-detection rules become less useful when the number of datapoints is very large since the likelihood increases of apparent patterns arising from randomness (false positives). 27 ) As is seen in figure 2, we failed to meet the KP1 target (80% within 7 days) in any week, and the (weighted) mean performance was 50%, varying widely (from 0% to 72% in a week). On KP2, the mean daily TaT was 8.9 days, again with a very wide set of process limits. ...
The Stem Cell Donation and Transplantation Department at NHS Blood and Transplant (NHSBT) facilitates unrelated donor haematopoietic stem cell transplantations for patients with life-threatening haematological malignancies or other blood diseases. Donors must be screened for infectious disease markers (IDMs) prior to donation. The purpose of IDM testing is to assess whether the donor currently has, or previously had, an infectious disease that could be transmitted to the recipient. The turnaround time (TaT) from sample collection to the return of IDM results is important to transplant clinicians and their patients. NHSBT has a target TaT of 80% within seven calendar days. Our initial analysis showed us that we failed to meet this in any week in the previous year, and our service was neither efficient nor consistent, so there was considerable improvement potential.
This quality improvement (QI) project aimed to improve the TaT of the IDM reporting service. We tested three change ideas through four Plan-Do-Study-Act (PDSA) cycles. We collected data on TaTs from our laboratory information management system (LIMS) and updated our statistical process control charts after each PDSA cycle. Over the course of the project, we reduced the mean TaT from 8.9 days to 5.5 days and increased the proportion of samples reported within the 7-day benchmark from 50% to 89%, reaching the key performance indicator (KPI) target.
Conducting this project was a rewarding experience. Although we encountered unanticipated technical issues during PDSA experiments, and we found that some change plans were not as effective in improving the KPIs as we expected, the improvement by the end of the study period was substantial. This QI project enabled us to meet our TaT targets and, ultimately, help ensure that our patients receive timely transplants. It suggests that QI may have wider applications across our part of NHSBT.
... In the specific case of the positive shift suggested at batches 47-49, unfortunately a sudden new increase in mortality appeared from batch 52 showing again an out of control process. In fact, the out of control point may exist without prior alarm or warning signal (46). In a real-time application of SPC charts, considering that every out of control point suggests that a problem impacts the process, it is appropriate to identify the problem and address it. ...
... Usually, signals for which no explanation can be found should not be regarded as poor performance by the model, but they rather might denote a problem which was not well realized by the caretaker (47). However, some "false alarm" might occur, for example when the SPC charts produce a signal of special cause variation but no following out of control points appear (46). In this study, several signals of special cause variation preceded one or more out of control points (Figure 1, batches 60 in MR chart; 42-49 in I chart), suggesting that these signals should be considered as "real alarms". ...
Oedema disease (ED) caused by Shiga-toxin-producing E. coli in pigs is a serious life-threatening disease, particularly among weaned piglets. When a preventive protocol is adopted in a specific farm, interpretation of effectiveness is often complicated in field conditions due to natural or “common cause” variation. For this reason, in this study a Statistical process control (SPC) approach was used to retrospectively evaluate the application of an ED preventive protocol (lower protein diet, ad-libitum fiber, vaccination at 5 days of age) in an infected commercial piglets' weaning site. The analysis was established over a 9-years period (n = 75 consecutive batches; 1,800 weaners per batch) using mortality for each batch as the key parameter of health and production; the statistics and the control limits (mean ± 3-fold sd; UCL, upper control limit; LCL, lower control limit) were based on data from the first 28 batches (Period 1) before the onset of the first ED clinical signs. The charts allowed the detection of defined out of control batches (i.e., with mortality out of the intervention limits) from batch 29 ongoing, exploring a Period 2 (unstable production and ED clinical signs; 36 batches) and a Period 3 (application of the ED preventive protocol; 11 batches). Mortality evaluation using SPC revealed a production system defined under-control (mean moving range bar = 1,34%; UCL = 4,37%; LCL = 0%) during Period 1. During Period 2, charts lost the state of statistical control, as showed by several signals of special cause variation due to the ED outbreak. Period 3 was characterized again by a state of statistical control, where no signals of special cause variation was showed. In conclusion, the retrospective application of SPC charts in the present study was able to confirm the efficacy of an ED preventive protocol in reducing mortality in a piglets' weaning site. SPC charting is suggested as an useful tool to provide insights into relationships between health, managerial, and welfare decision and some selected iceberg parameters in livestock.
... Unit-level changes in RAI-MDS 2.0 quality indicators are presented using statistical process control (SPC) charts [48]. Data were not normally distributed and thus the following SPC zones were created using pre-SCOPE (January, 2013 to December 2016) data: (a) zone − 3 = 1st-5th percentile; (b) zone − 2 = 5th-34th percentile; (c) zone − 1 = 34th-50th percentile; (d) zone + 1 = 50th-66th percentile; (e) zone + 2 = 66th-95th percentile; (f ) zone + 3 = 95th-99th percentile. ...
... Data were not normally distributed and thus the following SPC zones were created using pre-SCOPE (January, 2013 to December 2016) data: (a) zone − 3 = 1st-5th percentile; (b) zone − 2 = 5th-34th percentile; (c) zone − 1 = 34th-50th percentile; (d) zone + 1 = 50th-66th percentile; (e) zone + 2 = 66th-95th percentile; (f ) zone + 3 = 95th-99th percentile. SPC charts allow us to assess changes in processes or outcomes with time, and assume that in 'null effect' scenarios, data will be randomly distributed around a measure of central tendency [48]. Following the SPC Western Electric rules [49], non-random variation was detected if (a) one or more data points during the SCOPE pilot were beyond zone 3 of pre-SCOPE results, (b) two of three successive data points were beyond zone 2, or (c) four of five successive data points were beyond zone 1. ...
Background
Nursing home residents require daily support. While care aides provide most of this support they are rarely empowered to lead quality improvement (QI) initiatives. Researchers have shown that care aide-led teams can successfully participate in a QI intervention called Safer Care for Older Persons in Residential Care Environments (SCOPE). In preparation for a large-scale study, we conducted a 1-year pilot to evaluate how well coaching strategies helped teams to enact this intervention. Secondarily, we measured if improvements in team cohesion and communication, and resident quality of care, occurred.
Methods
This study was conducted using a prospective single-arm study design, on 7 nursing homes in Winnipeg Manitoba belonging to the Translating Research in Elder Care research program. One QI team was selected per site, led by care aides who partnered with other front-line staff. Each team received facilitated coaching to enact SCOPE during three learning sessions, and additional support from quality advisors between these sessions. Researchers developed a rubric to evaluate how well teams enacted their interventions (i.e., created actionable aim statements, implemented interventions using plan-do-study-act cycles, and used measurement to guide decision-making). Team cohesion and communication were measured using surveys, and changes in unit-level quality indicators were measured using Resident Assessment Instrument-Minimum Data Set data.
Results
Most teams successfully enacted their interventions. Five of 7 teams created adequate-to-excellent aim statements. While 6 of 7 teams successfully implemented plan-do-study-act cycles, only 2 reported spreading their change ideas to other residents and staff on their unit. Three of 7 teams explicitly stated how measurement was used to guide intervention decisions. Teams scored high in cohesion and communication at baseline, and hence improved minimally. Indicators of resident quality care improved in 4 nursing home units; teams at 3 of these sites were scored as ‘excellent’ in two or more enactment areas, versus 1 of the 3 remaining teams.
Conclusions
Our coaching strategies helped most care aide-led teams to enact SCOPE. Coaching modifications are needed to help teams more effectively use measurement. Refinements to our evaluation rubric are also recommended.
... RAI-MDS 2.0 quality indicators were calculated at the unit-level using quarterly data collected during the pilot, using statistical process control (SPC) methods [48]. Data were not distributed normally and thus the following SPC zones were created using pre-SCOPE (January, 2013 to December 2016) data: a) zone -3=1 st -5 th percentile; b) zone -2=5 th -34 th percentile; c) zone -1=34 th -50 th percentile; d) zone + 1=50 th -66 th percentile; e) zone + 2=66 th -95 th percentile; f) zone + 3=95 th -99 th percentile. ...
Background: Nursing home residents require daily support. While care aides provide most of this support they are rarely empowered to lead quality improvement (QI) initiatives. A previous proof of principle study, called Safer Care for Older Persons in Residential Care Environments (SCOPE), demonstrated that care aide-led teams can successfully participate in QI interventions. In preparation for a large-scale study, this one-year pilot evaluated how well the bundle of SCOPE coaching strategies helped care-aide led teams to enact these interventions. A secondary aim was to determine if improvements in resident quality of care occurred.
Methods: Using a modified IHI Breakthrough Collaborative Series model in a prospective single-arm study design, we randomly sampled 7 nursing homes in Winnipeg, Manitoba from the longitudinal Translating Research in Elder Care (TREC) cohort. Each SCOPE team had 5-7 front-line staff led by care aides. Teams received coaching to enact the intervention (i.e., to create actionable aim statements, implement QI interventions using plan-do-study-act [PDSA] cycles, use measurement to guide decision making) during three learning congresses, networked and shared learning experiences during these sessions, and received additional support from quality advisors between congresses. We used self-report data to code intervention enactment (‘poor’, ‘adequate’, ‘excellent’), and also measured improvement in team cohesion and communication. Secondarily, we observed changes in unit-level quality indicators using RAI-MDS 2.0 data.
Results: Most teams successfully enacted SCOPE. Five of 7 teams created adequate-to-excellent aim statements throughout the pilot (e.g., statements were specific, measurable, time-bound). While 6 of 7 teams successfully implemented PDSAs, only 2 reported spreading their idea to involve more than a few residents and/or staff on their unit. Three of 7 teams explicitly stated how measurement was used to guide decisions. Team cohesion and communication scored high at baseline, and hence improved minimally. Resident quality indicators improved in 4 of the 7 nursing home units.
Conclusions: Our bundled coaching strategies helped most care aide-led teams to enact SCOPE. Coaching modifications are needed in follow-up studies to help teams more effectively use measurement, and to spread successful interventions within the unit. More detailed and robust approaches are also needed to monitor treatment enactment.
... Given that routine surveillance is increasingly important in the control of economically important pathogens, producers should consider active, continuous analysis of surveillance results using statistical process control (SPC) or similar methods to provide context, determine the relevance of the diagnostic results, and thus characterize them as stable or unstable. The process of survey data analysis leads to a more comprehensive understanding of test performance and interpretation (Anhøj and Wentzel-Larsen, 2018). ...
Distinct from tests used in diagnostics, tests used in surveillance must provide for detection while avoiding false alarms, i.e., acceptable diagnostic sensitivity but high diagnostic specificity. In the case of the reproductive and respiratory syndrome virus (PRRSV), RNA detection meets these requirements during the period of viremia, but antibody detection better meets these requirements in the post-viremic stage of the infection. Using the manufacturer's recommended cut-off (S/P ≥ 0.4), the diagnostic specificity of a PRRSV oral fluid antibody ELISA (IDEXX Laboratories, Inc., Westbrook, ME USA) evaluated in this study was previously reported as ≥ 97%. The aim of this study was to improve its use in surveillance by identifying a cut-off that would increase diagnostic specificity yet minimally impact its diagnostic sensitivity. Three sample sets were used to achieve this goal: oral fluids (n = 596) from pigs vaccinated with a modified live PRRSV vaccine under experimental conditions, field oral fluids (n = 1574) from 94 production sites of known negative status, and field oral fluids (n = 1380) from 211 sites of unknown PRRSV status. Based on the analysis of samples of known status (experimental samples and field samples from negative sites), a cut-off of S/P ≥ 1.0 resulted in a diagnostic specificity of 99.2 (95% CI: 98.8, 99.7) and a diagnostic sensitivity of 96.5 (95% CI: 85.2, 99.2). Among 211 sites of unknown status, 81 sites were classified as antibody positive using the manufacturer's cut-off; 20 of which were reclassified as negative using a cut-off of S/P ≥ 1.0. Further analysis showed that these 20 sites had a small proportion of samples (18.0%) with S/P values just exceeding the manufacturer's cut-off (x̄ = 0.5). Whereas the remainder of positive sites (n = 61) had a high proportion of samples (76.3%) with high S/P values (x̄ = 6.6). Thus, the manufacturer’s cut-off (S/P ≥ 0.4) is appropriate for diagnostic applications, but a cut-off of S/P ≥ 1.0 provided the higher specificity required for surveillance. A previously unreported finding in this study was a statistically significant association between unexpected reactors and specific production sites and animal ages or stages. While beyond the scope of this study, these data suggested that certain animal husbandry or production practices may be associated with non-specific reactions.
