Systematic review of the use of Statistical Process Control methods to measure the success of pressure ulcer prevention

Abstract

Successful prevention of pressure ulcers is the end product of a complex series of care processes including, but not limited to, the assessment of vulnerability to pressure damage; skin assessment and care; nutritional support; repositioning; and the use of beds, mattresses, and cushions to manage mechanical loads on the skin and soft tissues. The purpose of this review was to examine where and how Statistical Process Control (SPC) measures have been used to assess the success of quality improvement initiatives intended to improve pressure ulcer prevention. A search of 7 electronic bibliographic databases was performed on May 17th, 2017, for studies that met the inclusion criteria. SPC methods have been reported in 9 publications since 2010 to interpret changes in the incidence of pressure ulcers over time. While these methods offer rapid interpretation of changes in incidence than is gained from a comparison of 2 arbitrarily selected time points pre‐ and post‐implementation of change, more work is required to ensure that the clinical and scientific communities adopt the most appropriate SPC methods.

1. DESCRIPTION OF THE HEALTH PROBLEM

Successful prevention of pressure ulcers is the end product of a complex series of care processes including, but not limited to, the assessment of vulnerability to pressure damage; skin assessment and care; nutritional support; repositioning; and the use of beds, mattresses, and cushions to manage mechanical loads on the skin and soft tissues.1 At least 25 quality improvement (QI) interventions have been adopted in academic medical centres in the United States between 2007 and 2012 to help support and encourage clinicians engaged in pressure ulcer prevention,2 covering all 4 domains of the framework by Nelson et al3 for QI (staff, leadership, information and information technology, and performance and management). Padula et al2 reported that academic medical centres implemented a mean of 7.5 of the 25 QI interventions in 2007 and a mean of 12.5 in 2012, with the increase associated with the decision not to reimburse hospitals for any costs associated with categories III and IV hospital‐acquired pressure ulcers in 2008. This rise in the implementation of QI initiatives around pressure ulcer prevention occurred across all domains of QI activity, with the greatest increases seen in leadership initiatives and the lowest rise in performance and improvement interventions.

While the increased emphasis on QI initiatives to support pressure ulcer prevention is encouraging, questions remain regarding how best to report their success in terms of fewer numbers of patients with pressure ulcers. Pre‐ and post‐implementation data on the occurrence of pressure ulcers have frequently been used to report success. Traditional methods, such as the analysis of pre‐ and post‐intervention measures, are prone to changes in the underlying population over time—for example, increased numbers of people at risk of pressure ulcers in the before or after population and the arbitrary selection of dates marking the “before” and “after” of an intervention. A further restriction of traditional methods lies in the time required to collect sufficient data to identify whether a change in outcome has been achieved.4

An alternative to arbitrary single measurements is the use of Statistical Process Control (SPC) tools5 allowing rapid, visual, longitudinal analyses of effectively real‐time changes in the number of patients with pressure ulcers, highlighting where the variation in incidence is simply a product of random change or marks an anomaly of unusually high or low pressure ulcer occurrence. SPC is a philosophy, a strategy, and a set of methods for the ongoing improvement of processes and systems to yield better outcomes. It is data‐based, and it has its foundation on the theory of variation.5

The application of SPC involves 3 main phases of activity:

1. Understanding the process and the specification limits.

2. Eliminating assignable (special) sources of variation so that the process is stable.

3. Monitoring the ongoing process, assisted using visual tools, to detect significant changes of variation.

The basic tool of SPC is the control chart that shows measurements plotted in time order, with 3 horizontal lines marking the central line (usually the mean) and 2 control limits (upper and lower) marking 3 times the standard deviation above and below the mean value.4 If a process is stable, then 99.73% of all measurements should fall within the bounds of the upper and lower control limits. Measurements that fall outside these limits are considered to mark a process not in control, with the introduction of a special‐cause variation, which signals the requirement for further investigation. Control charts can be based on limited sets of data with perhaps 20 to 30 measurements required to set the upper and lower control limits. Each subsequent measurement can be plotted on the control chart, and processes not in control can be rapidly identified.

Other visual charts are also used in SPC—for example, the run chart plots measurements in time order but does not include upper or lower control limits, effectively giving a visual indication of trends over time without the benefit of understanding whether the process is in control or not. Another version of the control chart is the funnel plot,6 where the control limits narrow as the sample size increases.

SPC is a well‐established method of managing complex processes,7 and the use of SPC has been widely reported in health care.8 In summary, it is a graphical technique used to identify the causes and monitoring of variation. It is particularly useful when monitoring a process to determine whether process changes have been effective. The purpose of this review is to examine where and how SPC has been used to assess the success of QI improvements intended to improve pressure ulcer prevention. This review will consider the use of control charts, run charts, and funnel plots to provide a visual representation of processes in control and where special‐cause variation exists.

2. RESEARCH QUESTIONS

The questions that this systematic review considered were:

1. What evidence is available to indicate that Statistical Process Control methods can provide insights into the success of pressure ulcer prevention?

2. Do Statistical Process Control methods offer greater insights into pressure ulcer prevention than measures collected before and after the implementation of quality improvement processes?

3. IDENTIFICATION OF STUDIES AND SEARCH STRATEGY

This review was commissioned by a manufacturer (Direct Healthcare Services Ltd, Caerphilly, Wales, UK), and no review protocol was developed or made available for comment.

The search strategy was as follows:

A search of 7 electronic bibliographic databases (Allied and Complementary Medicine Database (AMED), Database of Abstracts of Reviews of Effects (DARE), Cochrane Central Register of Controlled Trials (CCTR), NHS Economic Evaluation Database (NHS EED), WHO International Clinical Trial Registry Platform, Medline, and CINAHL Plus) was performed on May 17th, 2017, for studies that met the inclusion criteria. The databases were searched from inception to the date of the search, with no limitation on publication date. The review was limited to publications in the English language. The search strategy used in CINAHL Plus is described below:

S1 (MH “Pressure Ulcer”)

S2 TX pressure sore

S3 TX pressure injury

S4 TX decubitus ulcer

S5 S1 OR S2 OR S3 OR S4

S6 TX statistical process control

S7 TX SPC

S8 TX run charts

S9 S6 OR S7 OR S8

S10 S5 AND S9

Bibliographies of included studies were searched for further relevant studies. Hand searching of a non‐indexed journal (Wounds UK) was undertaken from volume 6(1) published in 2010 to the most recent issue in May 2017 (volume 13(2)). References were managed using EndNote version 17 (Thomson Reuters, Toronto, Canada).

3.1. Inclusion criteria

The inclusion criteria for the systematic review are summarised in Table 1.

Table 1.

Inclusion criteria

Questions	Criteria	Specification	Notes
1 and 2	Population	Patients at risk of developing pressure ulcers and patients with existing pressure ulcers
1 and 2	Intervention	Any quality improvement process implemented to improve pressure ulcer prevention	Secondary analyses of data collected during quality improvement processes were included in the review
1 and 2	Comparator	Pressure ulcer prevention performed prior to quality improvement intervention
1	Outcome	Use of Statistical Process Control methods to interpret changes in pressure ulcer incidence and occurrence	Use of P, T, and U charts to identify special cause variation around pressure ulcer occurrence
2	Outcome	Use of Statistical Process Control methods along with before and after comparisons of the effect of quality improvement processes	Use of P, T, and U charts compared with measures taken before and after the implementation of quality improvement processes.
1 and 2	Setting	Primary and secondary care	No restriction on geographical location
1 and 2	Study design	Systematic reviews, randomised and non‐randomised, cohort, case series and case studies, observational and qualitative studies	Publications with no primary data were excluded
1 and 2	Language	Restricted to English language publications

3.2. Exclusion criteria

Non‐English language and publications that did not report primary data were excluded; there were no further exclusion criteria regarding study design.

3.3. Study selection

Based on the above mentioned inclusion criteria, papers were selected for review from the titles and abstracts generated by the search strategy. This was performed independently by 2 reviewers (M.C., T.Y.), with discrepancies resolved by discussion. Retrieved papers were reviewed and selected against the inclusion criteria by the same independent process.

3.4. Data extraction

Data were extracted from included studies by 1 reviewer (M.C.) using a standardised data extraction form and were checked by a second reviewer (T.Y.). Data were gathered on the design, data source, context, sample size and characteristics, outcome measures, and results.

3.5. Critical appraisal—assessing risk of bias

QI studies were assessed for internal and external validity using the Quality Improvement Minimum Quality Criteria Set (QI‐MQCS).9 Additional notes were made regarding the SPC charts—were the methods for setting the upper and lower control limits explicitly stated in the publication and were the rules marking special cause variation outlined in each paper?

4. METHODS FOR ANALYSIS AND SYNTHESIS

4.1. Analysis and synthesis

No formal analysis of the data reported in the included studies was undertaken given the heterogeneous nature of the study populations, care settings, and reported outcomes. All studies had a narrative synthesis.

5. RESULTS

5.1. Number and type of studies included

The search found 18 records with abstracts after removal of duplicate records (Figure 1). Four records were excluded through abstract screening—2 were not in English,10, 11 and the other 2 12, 13 did not report SPC methods. The full text of 14 records were retrieved, with 5 excluded from the review (no primary data (n = 2)14, 15 and commentaries on a study included in the review (n = 2),16, 17 with the final exclusion being a report of the increased use of QI processes to improve pressure ulcer prevention following changes in reimbursement practices in the United States.2

Figure 1.

Flow chart showing records identified, screened, and included in the synthesis

5.2. Quality of studies—study characteristics and risk of bias

Table 2 summarises the 9 included studies,18, 19, 20, 21, 22, 23, 24, 25, 26 while Table 3 details the critical appraisal of the included studies using the QI‐MQCS tool. Of the 9 included studies, 3 were not studies of QI processes (1 secondary analysis of pressure ulcer prevalence surveys,19 1 a comparison of data presented at unit and facility levels,21 and the final paper a comparison of the use of regression analysis and SPC),24 and these 3 publications were not assessed using the QI‐MQCS tool. The assessed studies failed to identify the standard of care provided prior to the quality improvement intervention, while compliance with the intervention and the ability to spread or replicate the intervention were frequently not reported. None of the included studies reported the use of funnel plots.

Table 2.

Characteristics of the included studies

Reference	Title	Study design	Data source	Context	Sample size and characteristics	Outcomes measured	Results
Chaboyer et al,18	Transforming care strategies and nursing‐sensitive patient outcomes	Observational, time series design	Clinical incidents reported by ward staff (web‐based reporting form) and verified by nurse unit managers	2 medical units of a single Australian community hospital. Units had 35 and 30 beds and 26 FTE nursing staff. Data collected for 15 mo prior to introduction of “Transforming care at the bedside” initiatives and for 18 mo after implementation	Unspecified, data reported as mean proportion of reported errors that reportedly resulted in harm	Medication errors; Patient falls Pressure ulcers	Overall mean proportion of reported pressure ulcers that reportedly resulted in harm over the duration of the study was 66.7%. Wide variation throughout the study period; no evidence of systematic change being observed. Before and after quality improvement interventions; proportion of reported pressure ulcers that apparently resulted in harm fell from 91.3% to 46.6%, this variation likely occurred by chance given wide month‐to‐month variability over the study period.
Kottner and Halfens19	Using Statistical Process Control for Monitoring the Prevalence of Hospital‐acquired Pressure Ulcers	Observational, time series design	Annual point prevalence pressure ulcer surveys with consistent methodologies. Data analysed from surveys in 1998 to 2008	Hospitals that participated in the survey from 1998 to 2008.	6 hospitals with 11 444 patients considered at risk of developing pressure ulcers (by hospital; number of patients ranged from 1391 to 2382). All patients at risk of pressure ulcers had some impairment of their mobility (Braden score mobility sub‐scale scores of 1‐3 included).	Nosocomial Pressure ulcers with broken skin (EPUAP classification categories 2‐4)	Chi‐square trend tests suggested pressure ulcer prevalence fell significantly between the first and second half of the study period in 4/6 hospitals. SPC charts suggested common cause variation with only 1 data point in 1 hospital (2001 data) above the upper control limit.
Unbeck et al20	Design, application and impact of quality improvement “theme months” in orthopaedic nursing: A mixed method case study on pressure ulcer prevention	Retrospective mixed method case study with time series analyses	Retrospective record data on pressure ulcer risk assessment and prevalence	Single orthopaedic department (52 beds) in a suburban university hospital 1 d point prevalence data per month for 46 mo (January 2007 to October 2010). Patients only counted during the month of their admission to the department.	Each monthly prevalence survey included between 28 and 66 admissions. Intervention theme month November 2007 with data collected for 10 mo prior to theme month and for 3 y after.	Percentage of patients with a documented risk assessment Percentage of patients with documented pressure ulcers (severity unreported)	Prior to the interventions, 1.5% of patients had a documented risk assessment; after theme month, 48.3% assessed at admission. January to October 2007, 11.5% patients with pressure ulcers; 2010, 7.3%; stated as a statistically significant reduction. SPC mentioned in the 4 section but no display of control charts or discussion of common cause variation.
Norton et al21	Facility versus unit‐level reporting of quality indicators in nursing homes when performance monitoring is the goal.	Calculation of adjusted Resident Assessment Instrument—Minimum Data Set 2.0 indicators for the prevalence of category II to IV pressure ulcers. Other indictors—pain and prescription of antipsychotic drugs with no diagnosis of psychosis. Comparison of the indicators over time at either a unit or facility level	Risk‐adjusted calculated quality indicators calculated by the authors.	30 urban nursing homes with 94 units (mode number of units per home 3, maximum 8). Specific rules developed to guide the interpretation of the control charts.	25 homes with 2 or more units. Homes stratified into those with 2, 3, or 4 or more units and 2 homes from each strata; randomly selected data presented on 6 nursing homes. 13 to 18 observations from mid‐2007 to end 2011.	Prevalence of category II to IV pressure ulcers.	Facility‐wide run charts showed no special cause variation over the 4.5 y of follow up. Individual units showed improvement or deterioration.
Fletcher et al22	Real‐world evidence from a large‐scale multisite evaluation of a hybrid mattress	Retrospective survey of pressure ulcer occurrence data and monthly admissions	Clinical records of pressure ulcer occurrence	8 acute care trusts in England	5580 beds across 8 sites with 4230 replaced with powered hybrid mattress. Education and awareness days also held across all sites.	New pressure ulcers developed over 6 mo prior to mattress installation and for 6 mo after installation.	P chart showing pressure ulcer incidence across 5 sites that provided at least 12 mo incidence data. Data shown from 19 mo prior to mattress installation to 41 mo post‐implementation. Incidence fell 5 mo and after 11 mo post‐mattress implementation. Overall reduction of 56% in number of pressure ulcers in 6 mo post‐implementation (t test) unclear time point used for before implementation comparison.
McGrath et al23	Implementing hybrid support surfaces: key components for a step change in ulcer prevention	Survey of changes in pressure ulcer occurrence; time series at 2‐mo intervals over 2 y (April 2014 to February 2016).	Clinical records of pressure ulcer occurrence	Acute care hospitals and community hospitals in England	2 acute care hospitals and an unspecified number of community hospitals	Pressure ulcer prevalence Incidence of avoidable pressure ulcers Incidence of avoidable category I and II pressure ulcers Number of category I and II pressure ulcers Each reported before and after introduction of a powered hybrid mattress and targeted education	Reduction in pressure ulcer prevalence 4 mo after introduction of mattress and education. Sustained to end of data collection. Interruption in incidence of avoidable pressure ulcers 4 mo after introduction of mattress and education. Sustained to end of data collection. Reduction in incidence of category I and II pressure ulcers 4 mo after introduction of mattress and education. Sustained to end of data collection. Reduction of number of category I and II pressure ulcers 4 mo after introduction of mattress and education. Sustained to end of data collection.
Padula et al24	Building information for systematic improvement of the prevention of hospital‐acquired pressure ulcers with statistical process control charts and regression.	Regression analysis of retrospective hospital billing data and SPC charts of prospective patient records	Quarterly incidence of pressure ulcers 2004 to 2007 from hospital billing database. Data organised longitudinally by ward and quarter. Prospective patient record review for 2004 to 2007 on a single ward	4 surgical wards within a tertiary care facility	337 hospital‐acquired pressure ulcers from population of 43 844 (all 4 surgical wards)	Hospital‐acquired pressure ulcers Probit and logit Regression analyses control for age, age,2 length of stay, gender, medical history, anatomical location of PU. Incidence by ward controlled for age, age,2 length of stay, gender, and medical history. SPC charts for single ward—P chart incidence of hospital‐acquired pressure ulcers Comparison between incidence in facility and national incidence (7%) using t test. T chart number of days occurring between the last 25 hospital‐acquired pressure ulcers in 2007. P chart reliability of Braden risk assessment and documentation over 20 wk.	Regression analyses. Probit regression model with higher pseudo R 2. Model included length of stay, male gender, age and age,2 and interaction term male × age Panel data analysis. Hospital‐acquired pressure ulcer incidence ranged from 0% to 13% per quarter, no statistically significant differences between wards over time for age, male proportion, length of stay, or medical history. For each additional day in hospital, likelihood of developing a pressure ulcer increased by 0.28%. SPC P chart incidence. Incidence remained in statistical control. Incidence lower than national average (1.17% vs 7%) T chart of days between last 25 successive patients developing pressure ulcers. Mean time between pressure ulcers was 13.25 d, no special cause variation. There was a 57‐d interval between 16th and 17th events; might be micro‐level improvement in pressure ulcer prevention not indicated by special cause variation. 20‐wk observation of Braden score documentation. 93% risk stratified upon admission, process in statistical control except for 1 wk where completion fell below 70%.
Burston et al25	The effect of transforming care initiative on patient outcomes in acute surgical units: a time series study.	Cohort study with historical controls. Interrupted time series.	Data from hospital administrative database.	2 acute care surgical wards (28 beds, approximately 30 FTE staff, 20 Registered Nurses) in 450‐bed regional teaching hospital	Unit 1. Pre‐intervention n = 1945, post‐implementation n = 1805. Unit 2. Pre‐implementation n = 1613, post‐implementation n = 1762. 12 interventions implemented in Unit 1, 11 in Unit 2 implemented over 3‐mo period. Data gathered July 2008 to December 2010	Inpatient falls Hospital acquired pressure ulcer (I to IV, WHO ICD‐10‐AM L89 (L89.0‐L89.9)	Mean number of pressure ulcers per month Unit 1, 0.4%, Unit 2, 1.0% SPC P charts·3 SD used to calculate upper and lower control limits, Unit 1 in control throughout, Unit 2 special cause variation in 1 mo pre‐intervention; post‐implementation 7 consecutive points lower than mean
Jull et al26	Measuring hospital‐acquired pressure injuries: A surveillance programme for monitoring performance improvement and estimating annual prevalence.	Random sample of patients per month. Patients assessed by ward staff	Patients assessed by ward staff—presence/absence of risk assessment, pressure ulcer (present/absent), category and location, age, gender, ethnicity.	Teaching hospitals providing services for adults and children. Excludes acute mental health units, emergency departments, and delivery suites. Sample of 3 patients on wards with fewer than 11 beds, 7 on wards with 11 to 30 beds, and 14 on wards with over 30 beds.	Audits in March 2012 to February 2015 (32 259 eligible patients). 8274 patients audited, 517 with hospital‐acquired pressure ulcers. Intervention A+ SKIN‐E over 3 y with no specific intervention date. Risk assessment in use (A+), appropriate support surface (S), appropriate manual repositioning (K), appropriate continence management (I). Appropriate assessment and referral for malnutrition (N) and Patient and family education (E)	P charts (3 SD used to calculate upper and lower control limits). Hospital‐acquired pressure ulcers per month (including category I)	2 mo out of statistical control before first wave of intervention (implementation of risk assessment and A + SKIN‐E) then 9 successive months above mean level. 1 mo below lower control level 1.5 y post implementation. Step change 6 to 7 mo post implementation with reduced mean prevalence. Step Change identified by Maximum Likelihood Estimate

Table 3.

Quality Improvement study—Quality Improvement Minimum Quality Criteria Set (QI‐MQCS)

Study	Intervention	Organisational motivation	Intervention rationale	Intervention description	Organisational characteristics	Implementation	Study design	Comparator
Chaboyer et al18	Transforming care at the bedside	Not met	Met	Not met	Met	Met	Met	Not met
Unbeck et al20	Improvement theme months	Met	Met	Met	Met	Met	Met	Not met
Fletcher et al22	New support surface	Met	Met	Met	Not met	Met	Met	Not met
McGrath et al23	New support surface and education	Met	Met	Met	Met	Met	Not met	Not met
Burston et al25	Transforming care at the bedside	Met	Met	Met	Met	Met	Met	Not met
Jull et al26	Skin bundle	Met	Met	Not met	Met	Not met	Met	Not met

Study	Intervention	Data source	Timing	Adherence/fidelity	Health outcomes	Organisational readiness	Penetration/reach	Sustainability	Spread	Limitations
Chaboyer et al18	Transforming care at the bedside	Met	Met	Not met	Met	Met	Not met	Met	Not met	Met
Unbeck et al20	Improvement theme months	Met	Met	Met	Met	Met	Met	Met	Met	Met
Fletcher et al22	New support surface	Not met	Met	Not met	Met	Not met	Not met	Not met	Not met	Not met
McGrath et al23	New support surface and education	Met	Met	Not met	Met	Met	Met	Met	Not met	Not met
Burston et al25	Transforming care at the bedside	Met	Met	Not met	Met	Met	Not met	Met	Not met	Met
Jull et al26	Skin bundle	Met	Not met	Not met	Met	Met	Met	Not met	Not met	Met

The limited but growing literature reporting the use of SPC to interpret changes in pressure ulcer incidence was often unclear on how the upper and lower control chart limits were set (Table 4).

Table 4.

Quality of reporting of Statistical Process Control charts

Study	Explicit statement of upper and lower control limit	Identification of rules for specifying special cause variation
Chaboyer et al18	Yes	Yes
Kottner and Halfens19	No	Yes
Unbeck et al20	No	No
Norton et al21	No	Yes (on‐line supplementary file)
Fletcher et al22	No	Yes
McGrath et al23	No	No
Padula et al24	No	No
Burston et al25	Yes	Yes
Jull et al26	Yes	Yes

5.3. Discussion

SPC methodology has been used to interpret longitudinal data on pressure ulcer occurrence in 9 publications since 2010. The increasing use of SPC to rapidly identify whether changes occurred after the implementation of QI processes offers a new approach to the identification of clinically relevant temporal changes such as reductions in pressure ulcer incidence. However, consensus regarding how SPC is best applied to tracking changes in pressure ulcer incidence has not yet been reached. There was no agreement over the number of data points required pre‐ and post‐implementation to determine whether the processes resulting in pressure ulcer development were in control. The commentaries16, 17 after the publication of Unbeck et al20 highlight fundamental challenges such as the selection of the most appropriate control charts, particularly where the denominator at each time point, is liable to differ. Only 1 publication26 reported how apparent step changes in pressure ulcer incidence over time were identified.

Comparison of the interpretation of SPC charts and comparisons of pressure ulcer incidence at time points before and after intervention suggest that the use of SPC methods might provide more robust conclusions that can be drawn from single‐point estimates of incidence made at arbitrary time points19. Kottner and Halfens19 noted that single before and after prevalence measures suggested there had been a significant reduction in pressure ulcer prevalence over time in 4 of 6 hospitals that had participated in annual prevalence surveys over a 10‐year period. However, using SPC methods on the same dataset indicated that changes in pressure ulcer prevalence in 5 out of 6 hospitals had resulted from chance only. The paper concluded that “P charts provide more valuable information than single P values and are more helpful for monitoring institutional performance.” More generally, the use of SPC control and run charts allows health care improvement studies to rapidly assess whether processes are in control without delays in the identification of the stability of a process caused through building up sufficient data to perform robust before and after intervention comparisons.

5.4. Implications for health care and research

SPC tools are growing in importance in the health care arena and provide a tool‐kit for enabling organisations to promptly identify harm, waste, and variation. SPC tools can be used to evaluate current process performance, search for ideas for improvement, and inform if change has resulted in an improvement as well as offer a method to track implementation and, importantly, tell whether that improvement is sustainable.

Improvement takes place over time. Determining if improvement has happened and if it is lasting requires observing patterns over time. However, the potential value of SPC methodologies in health care will only be realised if the most appropriate methodologies are used to determine which control charts should be used to track processes over time. Decision trees and guidance are available to guide practitioners towards the correct methods to use to establish and interpret control charts, for example, References 27, 28, 29.

5.5. Limitations

The main limitation of this review was the exclusion of 2 publications not in the English language.

6. CONCLUSIONS

SPC methods have been reported in 9 publications since 2010 to interpret changes in the incidence of pressure ulcers over time. While these methods offer more robust interpretation of changes in incidence than is gained from a comparison of 2 arbitrarily selected time points pre‐ and post‐implementation of change, more work is required to ensure that the clinical and scientific communities use the most appropriate SPC methods. Regardless of the technical challenges of ensuring the appropriate use of SPC, such measures deserve wider adoption when considering complex processes such as pressure ulcer prevention.

Clark M, Young T, Fallon M. Systematic review of the use of Statistical Process Control methods to measure the success of pressure ulcer prevention. Int Wound J. 2018;15:391–401. 10.1111/iwj.12876