Recognising deterioration: nurses’documentation of vital signs–a systematic literature review

(1)

Degree project, 30 ECTS 2021-01-11

Recognising deterioration:

nurses’ documentation of vital signs

– a systematic literature review

Version 2

Author: Frida Collin, MB

School of Medical Sciences Örebro University

Örebro, Sweden

Supervisors: Prof. Elisabeth Hultgren-Hörnquist

School of Medical Sciences Örebro University, Sweden Dr. Nickcy N. Mbuthia Pwani University Kilifi, Kenya Word count Abstract: 248 Manuscript: 2855

(2)

Abstract

Introduction: Research show that patients frequently display abnormal vital signs as much

as 48h before a serious adverse event occur, such as cardiac arrest or unplanned intensive care unit admission. Therefore, early recognition of these changes trough vital sign

examination is essential in the prevention of deterioration. However, deterioration is often missed.

Aim: The aim was to investigate to what extent nurses in the general ward are documenting

vital signs prior to patient deterioration.

Methods: A systematic literature review was done using the databases PubMed and

CINAHL. Inclusion criteria: general ward and publication 2010-2020, exclusion criteria: emergency department, acute admission ward, paediatric ward, psychiatric ward,

interventions and continuous monitoring. Critical appraisal using tools from Joanna Briggs Institute. PRISMA statement for reporting of systematic reviews.

Results: Nine studies were included. It was seen that the fraction of cases who had vital signs

documented prior to deterioration was diverse, although never complete. Some studies showed an acceptable fraction of patients who were monitored in the hours prior to deterioration, but it was seen that the monitoring did not always escalate as the patient got worse. The vital signs most frequently documented were heart rate and pulse, though still missing in a large fraction of charts. Respiratory rate was documented less than the other vital signs.

Conclusions: This study suggests that documentation of vital signs prior to deterioration is

diverse but often incomplete. Further research is needed to understand what can be done to improve vital sign documentation on general wards.

(3)

Abbrevations

BP = Blood pressure

CINAHL = Cumulated Index to Nursing and Allied Health Literature EWS = Early Warning Score

HR = Heart rate

ICU = Intensive care unit

NEWS = National Early Warning Score

NICE = National Institute for Health and Care Excellence

PubMed = US National Library of Medicine National Institutes of Health RR = Respiratory rate

(4)

Introduction

Nurses are increasingly providing complex care for more acutely ill patients at general wards, as the population is ageing and the intensive care beds are scarce [1]. Unexpected

deterioration of these patients on general wards is a major concern, and a study set in four Finnish hospitals showed that 51% of 100 cardiac arrests occurred on general wards [2]. However, research show that signs of clinical deterioration can be found hours before an adverse event [3]. Examples of adverse events are unplanned intensive care unit (ICU) admission or cardiac arrest [4]. It is suggested that abnormal vital signs can present as much as 6 to 48 hours in advance of such an event [4–6].

In a study that reviewed 7.851 cases of in-hospital cardiac arrest it was found that 59% of the patients had at least one abnormal vital sign 1-4h prior to cardiac arrest and that 13% had at least one severely abnormal vital sign [7]. Hence, an established key in the detection of deterioration is the recording of vital signs by the ward nurse [8,9]. To detect deterioration many hospitals use track and trigger systems [10]. These can be single- or

multiple-parameters systems [11]. Examples of such systems are Cardiac Arrest Risk Triage, Modified Early Warning Score, Early Warning Score (EWS) and VitalPAC Early Warning Score which are most often comprised of 5-7 items. The most common parameters included in these tools are heart rate (HR), respiratory rate (RR), systolic blood pressure, temperature and mental status, but additional items such as decreased urinary output, O2 saturation, difficulty

breathing, increased supply of supplemental O2 and nurses’ concern are sometimes included

[12]. The four primary vital signs are defined in the MeSH-term Vital signs as RR, HR, blood pressure (BP) and temperature [13]. Track and trigger systems compose the afferent part of rapid response systems. When deterioration is detected by a track and trigger system, it activates a response, i.e. the efferent part of the rapid response system. The response consists of for example medical emergency teams, or critical care outreach. [14]

The optimal method for detecting a deteriorating patient remains unknown, but research suggest that improved vital sign monitoring could improve outcomes [15]. Early detection and management of deterioration have been shown to improve patient outcomes, and to reduce healthcare resource utilization [16]. Research suggest that 18-66% of cardiac arrests are potentially preventable [17–19]. However, the deteriorating patient is often missed and the abnormal vital signs misinterpreted or mismanaged [19]. Knowledge is still lacking in

(5)

this area, and this systematic literature review hope to illuminate areas for improvement regarding documentation prior to patient deterioration.

Aim

The aim was to investigate to what extent nurses in the general ward are documenting vital signs prior to patient deterioration. The hypothesis is that there is a minor lack in

documentation prior to deterioration.

The research questions were:

1) In the general ward, how are nurses documenting vital signs prior to patient deterioration?

2) Which primary vital signs are documented most frequently? 3) Which primary vital signs (if any) are missed to be documented?

Material and Methods

Design

This is a systematic literature review with a descriptive design, without a quantitative pooling of data (meta-analysis).

Search / research

The search was built using Boolean operators AND and OR together with both MeSH terms and free text search in title and abstract. To create a well-built search, extensive research was made by reading abstracts and keywords of articles on the subject studied. Several attempts with different search blocks were executed before the final search was made. In the final search, terms used were variations of 1) vital signs, 2) documentation and 3) deterioration. See Supplementary Table 1 and 2 for further details on the search. The databases searched were PubMed (US National Library of Medicine National Institutes of Health) and CINAHL (Cumulated Index to Nursing and Allied Health Literature).

Selection

Inclusion and exclusion criteria were used to narrow down the result to contain only relevant articles. Inclusion criteria were: general ward and publication 2010-2020, whereas exclusion

(6)

criteria were emergency department or acute admission ward, paediatric ward, psychiatric ward, interventions and continuous monitoring.

The limitation of the publication year was set because local routines and practices are continuously updated. As per protocol, review articles and duplicate articles were excluded. In the process of screening articles by title and/or abstract every article that could somehow include relevant findings were included however small the chance. No exclusion was made regarding language. Repeated abstracts presented at different congresses or abstracts published later as a full article by the same authors were excluded.

Critical appraisal / Scientific assessment

A standardized critical appraisal was done using templates from Joanna Briggs Institute (https://joannabriggs.org/ebp/critical_appraisal_tools) and the PRISMA statement for reporting of systematic reviews (http://www.prisma-statement.org/). Templates were used according to type of study. Articles were assessed for methodological quality and the extent to which a study addressed the possibility of bias in its design, conduct and analysis was determined. A study was considered high quality if all criteria were fulfilled, moderate if most criteria were fulfilled and low if it lacked in several areas. Studies deemed to be of low quality were excluded.

Ethical consideration

Seven out of the nine articles included in the final review had ethical approval. One study did not state if there was an ethical approval. The study that reported an absence of ethical

approval had approval from the data protection authority. No study obtained patient consent. When the need for consent was mentioned it was argued that it was waived as the majority of studies were retrospective and therefore did not risk influencing or disturbing ongoing care.

Results

The selection process

The search in PubMed generated 614 articles and the search in CINAHL generated 266 articles, (Supplementary Table 1 and 2). After removal of duplicates 684 articles remained. One article had a corrected version included as well, leaving it to a total of 683 unique records. The result of the selection process is shown in Figure 1. Two records were added

(7)

from PubMed after being cited in included articles and one record was added when found on PubMed reading articles for the background. When records were screened by title and/or abstract – if abstracts were missing or incomplete, articles were read in full text - 609 records were excluded, and 75 articles were thereafter read in full text.

When records were read in full text for eligibility, 63 records were excluded with reason. The reasons for exclusion were: Other setting than general ward, interventional study design, setting not defined, continuous data, no presentation of missing vital parameters, no

deterioration measured or defined and paediatric population. One record was a master thesis and thus excluded. Grey literature does not contribute substantially to the result of a literature review and since there was a sufficient amount of records included in this review the master thesis could be excluded [20] . Studies included were cohort, cross sectional, prevalence or case control studies.

Figure 1: The selection of articles was done by systematic exclusion of those that did not fit

(8)

Summary of the articles

Supplementary Table 3 summaries the key findings of the included studies. All the articles used adverse events as surrogate markers for deterioration in patients. Cardiac arrest was studied the most, as well as unplanned ICU admission. Other adverse events were: transfer from the ward to the emergency department, cardiac arrest team calls and unexpected death. Eight studies were of retrospective design and one study used a prospective design. Five studies stated that there was a track and trigger systems in place at the time of the study, one study stated that there was not, and two studies did not state its’ presence or absence. All but one article included the four primary vital signs and supplementary O2 in the analysis and

seven articles included mental state. Five studies excluded patients with Do Not Resuscitate orders, and one study excluded patients who had been hospitalised less than 24h. Six studies did not state local guidelines.

Documentations prior to patient deterioration

In a retrospective chart review of patients suffering from cardiac arrest it was reported that 50% of records did not reach minimum standard of practice in the 12h preceding an event. There was significantly poorer adherence on the weekend and bank holiday [21]. A case-control study showed that the fraction of patients who had a National Early Warning Score (NEWS) assessment prior to cardiac arrest was as following: 24-18h 44%, 18-12h 50%, 12-6h 53%, 6-0h 49%, for both cases and controls combined. However, this study did not present additional data showing number of patients who had vital signs documented more than once during the time reviewed. Fiftyone out of 212 patients were excluded from the study because their records lacked the 4/7 minimum parameters that were needed for calculation of NEWS. Out of the 970 NEWS measurements that were reviewed, 21% had missing data [22]. Another study reported that a mean of 2.3 (range 0-12) vital signs were documented 8h prior to an event [19]. The result was similar in a study where a mean of 3 out of 6 (range 0-5) possible vital parameters were documented on the nursing shift prior to unplanned ICU admission. For 9% of cases there were no parameter documented and for 3% there were 5 parameters documented [23]. No record was complete, which was also the finding in a case control study of postoperative death [24]. A prospective study, where minimum monitoring was ‘measurements of vital parameters every 12h’, reported that 88% of the cases had at least two sets of EWS recorded in the 24h preceding the event. When assessed against the escalation protocol the monitoring frequency was fully adhered to in 13% of the unplanned ICU admissions and in 27% of the CA and unexpected deaths. In total,

(9)

19% of cases met the minimal observation interval according to the protocol. Patients with higher EWS (meaning more abnormal vital signs) was significantly associated with a lower likelihood of being monitored according to the escalation protocol, the incidence falling from 83% for EWS ≤1 to 6% for patients with EWS ≥9 [25].

In a study comparing patients who had surveillance-oriented nursing diagnoses (SOND) documented in their chart and those that did not (NO SOND), it was seen that mean time between the last documented measurements and cardiac arrest was 2.8 and 4.9 hours, respectively (p<0.001) [26]. In the SOND group 75% had measurements documented at intervals shorter than 6h, compared to 16% in the NO SOND group (p<0.0001). 25% in the SOND group had measurements exceeding 6h intervals, compared to 84% in the NO SOND group (p<0.001). Patients (both groups) who had Assessment of Basic Physiological

Measurements documented at intervals shorter than 6h survived in 63% of cases, but only 38% of those who had longer intervals survived [26].

Vital signs documented most frequently

In the included articles it was found that out of the four primary vital signs, HR and BP were documented in the biggest fractions of charts (Table 1). Both HR and BP were documented in >90% of the charts in 3 articles. The fraction of documented HR ranged from 67% to 100%, and BP ranged from 71% to 100%. One article stated that there was a significantly higher fraction of documented HR and BP before an adverse event in cases that were deemed to be potentially preventable, than those that did not [19]. In the study that compared patients with SOND and NO SOND there was no significant differences between the groups in HR and BP documentation, which was 100% for both groups [26].

Vital signs documented that where missed to be documented

Of those studies that reported the frequency of vital parameters documentation, RR was the least documented primary vital parameter in all but one. RR documentation ranged from 0% to >90% in the included studies. One study reported that none of the patients had RR

documented in the 8h period prior to an event [19]. In a case control study the results was similar, where none of the cases had RR documented and 2% of controls [24]. There was significantly more documentation of RR in the SOND group than in the NO SOND group (p<0.001) [26].

(10)

Table 1. Documentation of the primary vital signs as fraction of cases in the included studies.

Studies that reported frequency of vital signs documentation

[21] [27] [22] [19] [23] [24] [28] [26]

Hours reviewed 12h 72h 24h 8h 1 nursing

shift 8h 24h 24h

Vital signs (%)

Heart rate > 90 67 85 100** / 98*** 73 100

Blood pressure > 90 71 88 100** / 100*** 84 100 Respiratory rate > 90 Least* 0 14 0** / 2*** 65 26 Temperature <70 82 65 69 100** / 96*** 80 100

*_{= did not report exact numbers,}**_{= case,}***_{= control}

Discussion

In this systematic literature review we found that the documentation of vital signs prior to deterioration was diverse, although it was never reported to be complete. Some studies showed an acceptable fraction of patients being monitored in the hours prior to deterioration, but it was seen that the monitoring did not always escalate as the patient got worse. At the same time, it was found that when nurses assigned a patient with a SOND, and thereby observing that there was a risk of deterioration, monitoring was subsequently done more frequently and records were more complete. The vital signs that were most frequently

documented were sometimes still missing in a large fraction of charts. It was evident that RR was documented less than the other vital signs.

The NICE (National Institute for Health and Care Excellence) Guideline 50 advises that standard physiological observation should be done at least every 12h, and with increased frequency when abnormal vital signs are detected [29]. This was not reported to be achieved in any of the studies, the closest being 88% of patients having NEWS documented twice in 24h. The number of incomplete records seen in this review could be due to many different reasons. It has previously been found that staffing levels affect the level of adherence to protocol [30]. It has also been stated that inadequate routines could hinder vital sign

documentation [31]. This could be a possible mechanism for our results, as the better part of studies did not state local guidelines and it stands to question if there were any implemented. At the same time, it has been suggested that vital signs measured also appears to rely on nurses’ judgment [32]. This could be realistic for the increased frequency of monitoring in patients showing signs of clinical deterioration or abnormal vital signs, but the minimum

(11)

monitoring frequency should be adhered to. A multitude of factors influence the clinical course of a critically ill patient, and the routine of observations on the wards to detect deterioration could be an unrealistic expectation to predict outcomes [33], as is suggested in this review. It is concerning that RR was the vital sign being documented least frequently, because it has previously been found to be the most important vital sign related to

deterioration [34] as well as being an early indicator of disease. [35] It has previously been reported that RR is neglected by ward staff [36] and this review emphasizes those findings. The completeness of vital sign documentation is required for appropriate use of rapid response systems, and it is thus of importance that it is done sufficiently.

One limitation of this systematic review is the use of only two databases, and it is possible that additional records eligible for inclusion could have been found in other databases. A broader search could be done by using additional databases. Another limitation is that a single researcher screened the titles and abstracts for eligibility, as opposed to the usual two that are normally required. Furthermore only two studies used controls, and it is therefore difficult to draw any conclusions as to whether documentation of vital signs was done more or less than in the average in-hospital patient, i.e. if signs of deterioration affected the monitoring. Lastly, documentary data may not reflect practice, and it is possible that vital signs are taken more often than are documented. In the case of a medical emergency, the focus might not be to document. However, there is a legal problem in cases of medical liability – “if it isn’t documented, it isn’t done”. Moreover, nurses have probably been in contact more often with patients than to take vital signs, for example administrating drugs or to attend a call. However, this is not synonymous with surveillance.

It is a strength that all studies used surrogate markers for deterioration, as it is a well-used way to measure the outcome that is otherwise difficult to define. Another strength of the results is that the majority of studies excluded Do Not Resuscitate orders, which could have exaggerated the results in either direction.

A question that is yet to be answered is why there was such a high frequency of incomplete records in the included studies. What is the ideal frequency for vital sign monitoring when taking both the safety of the patients as well as the nurses’ workload into account? As this review highlights the lack of case control studies on the subject, the difference between

(12)

Conclusions

This systematic literature review suggests that documentation of vital signs prior to

deterioration is diverse but often incomplete. Further research is needed to understand as to what an ideal frequency for vital sign documentation on general wards would be.

Acknowledgment

I would like to offer a special thanks to my supervisors Elisabeth Hultgren-Hörnquist and Nickcy N. Mbuthia. Thank you for flexibility and ingenuity when plans changed. I feel fortunate to have had you as my supervisors.

References

1. Steen CD, Costello J. Teaching pre-registration student nurses to assess acutely ill patients: an evaluation of an acute illness management programme. Nurse Educ Pract 2008;8:343–51.

2. Nurmi J, Harjola VP, Nolan J, Castrén M. Observations and warning signs prior to cardiac arrest. Should a medical emergency team intervene earlier? Acta Anaesthesiol Scand 2005;49:702–6.

3. Churpek MM, Yuen TC, Edelson DP. Risk stratification of hospitalized patients on the wards. CHEST 2013;143:1758–65.

4. Buist MD, Jarmolowski E, Burton PR, Bernard SA, Waxman BP, Anderson J.

Recognising clinical instability in hospital patients before cardiac arrest or unplanned admission to intensive care. A pilot study in a tertiary-care hospital. Med J Aust 1999;171:22–5.

5. Franklin C, Mathew J. Developing strategies to prevent inhospital cardiac arrest: analyzing responses of physicians and nurses in the hours before the event. Crit Care Med 1994;22:244–7.

6. Hillman KM, Bristow PJ, Chey T, Daffurn K, Jacques T, Norman SL, et al. Antecedents to hospital deaths. Intern Med J 2001;31:343–8.

7. Andersen LW, Kim WY, Chase M, Berg K--> M, Mortensen SJ, Moskowitz A, et al. The prevalence and significance of abnormal vital signs prior to in-hospital cardiac arrest. Resuscitation 2016;98:112–7.

8. Andrews T, Waterman H. Packaging: a grounded theory of how to report physiological deterioration effectively. J Adv Nurs 2005;52:473–81.

9. McBride J, Knight D, Piper J, Smith GB. Long-term effect of introducing an early

(13)

10. Wuytack F, Meskell P, Conway A, McDaid F, Santesso N, Hickey FG, et al. The effectiveness of physiologically based early warning or track and trigger systems after triage in adult patients presenting to emergency departments: a systematic review. BMC Emerg Med 2017;17:38.

11. Al-Qahtani S, Al-Dorzi HM. Rapid response systems in acute hospital care. Ann Thorac Med 2010;5:1.

12. Smith MEB, Chiovaro JC, O’Neil M, Kansagara D, Quiñones AR, Freeman M, et al. Early warning system scores for clinical deterioration in hospitalized patients: a systematic review. Ann Am Thorac Soc 2014;11:1454–65.

13. Vital Signs - MeSH - NCBI [Internet]. [cited 2020 Nov 23]; Available from: https://www-ncbi-nlm-nih-gov.db.ub.oru.se/mesh/?term=vital+signs

14. Lyons PG, Edelson DP, Churpek MM. Rapid response systems. Resuscitation 2018;128:191–7.

15. DeVita MA, Smith GB, Adam SK, Adams-Pizarro I, Buist M, Bellomo R, et al. “Identifying the hospitalised patient in crisis”--a consensus conference on the afferent limb of rapid response systems. Resuscitation 2010;81:375–82.

16. Khan NA, Quan H, Bugar JM, Lemaire JB, Brant R, Ghali WA. Association of postoperative complications with hospital costs and length of stay in a tertiary care center. J Gen Intern Med 2006;21:177–80.

17. Galhotra S, DeVita MA, Simmons RL. Mature rapid response system and potentially avoidable cardiopulmonary arrests in hospital. Qual Saf Health Care 2007;16:260–5. 18. Hodgetts TJ, Kenward G, Vlackonikolis I, Payne S, Castle N, Crouch R, et al. Incidence,

location and reasons for avoidable in-hospital cardiac arrest in a district general hospital. Resuscitation 2002;54:115–23.

19. De Meester K, Van Bogaert P, Clarke SP, Bossaert L. In-hospital mortality after serious adverse events on medical and surgical nursing units: a mixed methods study. J Clin Nurs John Wiley Sons Inc 2013;22:2308–17.

20. Hartling L, Featherstone R, Nuspl M, Shave K, Dryden DM, Vandermeer B. Grey

literature in systematic reviews: a cross-sectional study of the contribution of non-English reports, unpublished studies and dissertations to the results of meta-analyses in child-relevant reviews. BMC Med Res Methodol [Internet] 2017 [cited 2020 Nov 17];17. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5395863/

21. Odell M. Detection and management of the deteriorating ward patient: an evaluation of nursing practice. J Clin Nurs John Wiley Sons Inc 2015;24:173–82.

22. Spångfors M, Molt M, Samuelson K. In-hospital cardiac arrest and preceding National Early Warning Score (NEWS): A retrospective case-control study. Clin Med Lond Engl 2020;20:55–60.

(14)

24. Kyriacos U, Jelsma J, Jordan S. Record review to explore the adequacy of post-operative vital signs monitoring using a local modified early warning score (mews) chart to evaluate outcomes. PloS One 2014;9:e87320.

25. Petersen JA, Mackel R, Antonsen K, Rasmussen LS. Serious adverse events in a hospital using early warning score - what went wrong? Resuscitation 2014;85:1699–703.

26. Maria-Eulàlia Juvé-Udina, Núria Fabrellas-Padrés, Jordi Adamuz-Tomás, Sònia Cadenas-González, Maribel Gonzalez-Samartino, Laura de la Cueva Ariza, et al. Surveillance nursing diagnoses, ongoing assessment and outcomes on in-patients who suffered a cardiorespiratory arrest. Rev Esc Enferm USP 2017;51:1–8.

27. Correia N, Rodrigues RP, Sá MC, Dias P, Lopes L, Paiva A. Improving recognition of patients at risk in a Portuguese general hospital: results from a preliminary study on the early warning score. Int J Emerg Med 2014;7:22.

28. Stevenson JE, Israelsson J, Nilsson GC, Petersson GI, Bath PA. Recording signs of deterioration in acute patients: The documentation of vital signs within electronic health records in patients who suffered in-hospital cardiac arrest. Health Informatics J

2016;22:21–33.

29. Centre for Clinical Practice at NICE (UK). Acutely Ill Patients in Hospital: Recognition of and Response to Acute Illness in Adults in Hospital [Internet]. London: National Institute for Health and Clinical Excellence (UK); 2007 [cited 2020 Dec 21]. Available from: http://www.ncbi.nlm.nih.gov/books/NBK45947/

30. Redfern OC, Griffiths P, Maruotti A, Recio Saucedo A, Smith GB. The association between nurse staffing levels and the timeliness of vital signs monitoring: a retrospective observational study in the UK. BMJ Open 2019;9:e032157.

31. Stevenson JE, Israelsson J, Petersson G, Bath PA. Factors influencing the quality of vital sign data in electronic health records: A qualitative study. J Clin Nurs John Wiley Sons Inc 2018;27:1276–86.

32. Cardona-Morrell M, Prgomet M, Lake R, Nicholson M, Harrison R, Long J, et al. Vital signs monitoring and nurse–patient interaction: A qualitative observational study of hospital practice. Int J Nurs Stud 2016;56:9–16.

33. Morgan RJM, Wright MM. In defence of early warning scores. Br J Anaesth 2007;99:747–8.

34. Schein RM, Hazday N, Pena M, Ruben BH, Sprung CL. Clinical antecedents to in-hospital cardiopulmonary arrest. Chest 1990;98:1388–92.

35. Cretikos M, Chen J, Hillman K, Bellomo R, Finfer S, Flabouris A, et al. The objective medical emergency team activation criteria: a case-control study. Resuscitation

2007;73:62–72.

36. Cretikos MA, Bellomo R, Hillman K, Chen J, Finfer S, Flabouris A. Respiratory rate: the neglected vital sign. Med J Aust 2008;188:657–9.

(15)

APPENDIX 1 – SEARCH STRATEGY AND MESH TERMS

USED

Supplementary Table 1. Search 2020-10-05 13.00 in PubMed

Search blocks No. of hits

Vital parameters

1. ("Vital Signs"[Mesh]) OR ("vital sign*"[Title/Abstract]) 424 505

Documentation

2. (("Documentation"[Mesh]) OR ("Monitoring, Physiologic"[Mesh])) OR (document*[Title/Abstract])) OR (monitor*[Title/Abstract])) OR (record*[Title/Abstract])) OR (chart*[Title/Abstract])

3 197 725

Deterioration

3. ("Clinical Deterioration"[Mesh]) OR (deteriorat*[Title/Abstract]) 127 878

1 AND 2 AND 3

4. 1 225

Limiters - Published Date: In the last 10 years

5. 614

Supplementary Table 2. Search 2020-10-06 14.00 in CINAHL

Search blocks No. of hits

Vital parameters

1. (MH “Vital Signs+”) OR TI “vital sign*” OR AB “vital sign*” 24 076

Documentation

2. (MH "Charting") OR (MH "Coding") OR (MH "Medical Records") OR (MH "Monitoring, Physiologic+") OR TI ( document* OR monitor* OR record* OR chart* ) OR AB ( document* OR monitor* OR record* OR chart* )

604 386

Deterioration

3. (MH "Clinical Deterioration") OR TI deteriorat* OR AB deteriorat* 24 289

1 AND 2 AND 3

4. 333

Limiters - Published Date: 20101001-20201031

(16)

APPENDIX 2 - DATA COLLECTED

Supplementary Table 3. A summary of the key data collected from the included studies.

Author, year, country

Methods Vital

signs

Results Quality Ethical

approval /consent Odell, 2015, England [21]

Retrospective chart review against EWS of patients on

general wards suffering

cardio-respiratory arrest. Reviewed for 12h prior to event. Data collected on 11 standard elements, of which 7 were minimum standard of practice. BP, HR, temp, RR, SpO2.

123 events. In 20% all 11 elements were recorded. 50% of records did not reach minimum standard of practice, 29% were deemed basic.

HR, BP, RR and SpO2 documented in

>90% of records. Temp documented <70% of records. Weekend or bank holiday showed significantly poorer adherence.

Moderate Yes / Not stated Correia et al, 2014, Portugal [27] Retrospective calculation of EWS at -12, -24 and 72h prior to assistance by

outreach team, with

consecutive transfer to

emergency room.

Population consisted of the general ward patients with this event.

Not stated

65 patients. 18 patients excluded due to significant absence of vital sign records. RR least recorded.

Moderate Not stated / Not stated Spångfor s et al, 2020, Sweden [22]

Retrospective record review using a 1:2 matched case-control design of patients suffering cardiac arrest on

general ward. For

calculation of NEWS 4/7 parameters needed to be registered. SBP, HR, temp, RR, SpO2, MS, SuO2

51 patients without any NEWS measurements excluded. 127 patients with 254 controls generated 970 NEWS measurements. Of these 21% had missing data. Temp was missing for 18% of records and SuO2 for 10%. Fraction with NEWS assessment prior to CA: 24-18h 44%, 18-12h 50%, 12-6h 53%, 6-0h 49%. High Yes / No De Meester et al, 2013, Belgium [19]

Retrospective record review of deaths following cardiac

arrest team calls, or

unplanned ICU admissions

from general wards.

Recorded vital signs 8h prior to event reviewed. Observation was considered to have taken place when

one vital sign was

documented. SBP, HR, temp, RR, SpO2, MS, SuO2.

70 cases of cardiac arrest and 129 unplanned ICU admissions, of which 63% died. A mean of 2.3 (range 0-12) chart entries/vital signs) documented.

Fraction of charts which had

following vital signs documented:

67% HR, 71% SBP, 62% SpO2, 48%

SuO2, 5% MS, 65% temp, 0% RR. All but MS showed significant difference

between cases deemed to be

potentially preventable and not.

High Yes / Not

stated

Jonsson

et al,

2011,

Retrospective record review of vital parameters and MEWS prior to emergency

BP, HR, temp,

65 patients. Mean number of MEWS parameters documented 3/6 (range 0-5). For 9% no parameter documented,

Moderate Yes / Not stated

(17)

Iceland [23]

admission to ICU from

general ward. MEWS

composed of 6 parameters. Records reviewed during the nursing shift prior to emergency admission.

RR, MS, UO, SpO2

for 3% there were 5. No record had all 6 parameters documented.

RR documented for 14% of patients, UO 42%, temp 69%, SpO2 80%, HR 85%, BP 88%. Kyriacos et al, 2014, South Africa [24]

Retrospective record review using 1:4 case control design. Population: patients who died up to 7 days postoperatively. A local

MEWS developed and

calculated for 8h

postoperatively. No EWS in place. Surgical wards.

BP, HR, temp, RR, MS, SpO2, UO

11 cases fit the criteria. 4 cases excluded due to records unavailable. 44 controls. 7/11 cases died >8h after operation.

No patients all 7 parameters recorded. All patients had recordings for SBP. Zero cases had RR recorded, 2% of the controls. All cases had recordings for HR, 98% of the controls. 55% of cases had recordings of SpO2, 7% of

the controls (p<0.001). In case group 13 recordings for sat, 7 for the control group (p<0.0001). All cases had recordings for temp, 96% of controls. 36% of cases had recordings for MS, 68% of controls. 82% of cases had recordings for UO, 96% of controls.

High Yes / Institutiona l consent Stevenso n et al, 2016, Sweden [28]

Retrospective record review of patients who suffered cardiac arrest on general wards. Data collected for 24h prior to event. Local guidelines: BAS recorded at least once per day.

SBP, HR, temp, RR, SpO2, SuO2, MS

127 patients included. Fraction of patients who had vital parameters documented: RR 65%, BP 84%, HR 73%, SpO2 73%, temp 80%, MS 15%, SuO2 45%. Moderate Yes Juvé-Udina et al, 2018, Spain [26]

Retrospective record review of patients who suffered

cardiac arrest. Data

collected for 24h prior to event.

The population was divided based on existence of one or more SONDs in the records. ABPP in place at the time of study. Study period 4 years. Randomly selected sample.

BP, HR, temp, RR, SpO2, MS, UO, Mi, CG

450 cases. 58% of patients had SOND documented.

Fraction of patients had ABPP documented: 100% HR, 100% BP, 100% temp, 99% MS, 93% Mi, 90% SpO2, 76% UO, 26% RR, 51% CG.

Significance between the groups where the following was documented

Populärvetenskaplig sammanfattning

I takt med att världens befolkning lever allt längre, blir fler människor multisjuka. Detta avspeglar sig i vården, och det ställs idag högre krav på sjuksköterskor som övervakar multisjuka patienter på vårdavdelningar. Genom att kontrollera vitala tecken såsom tex blodtryck, andningsfrekvens, puls och kroppstemperatur kan en försämring av

allmäntillståndet ses tidigt, vilket gör det möjligt att förebygga allvarliga konsekvenser som tex hjärtstopp. Vi har samlat de tio senaste årens forskning på området.

Genom att titta på publicerad forskning såg vi att långt ifrån alla patienter på vårdavdelningar övervakas i tillräckligt hög utsträckning. Vitala tecken undersöks inte i enlighet med

rekommendationer och en stor andel av patienterna saknar en eller flera vitala tecken dokumenterade i sin journal timmarna innan hjärtstopp eller allvarlig försämring. Forskningen tyder också på att övervakning inte alltid ökar i intensitet i takt med att patienterna uppvisar tecken på försämring. Andningsfrekvens är det vitala tecken som dokumenteras allra minst i journaler och i två studier var det inte dokumenterat i en enda journal.

Det är viktigt att upptäcka patienter som försämras i tid. Intensivvården har få platser tillgängliga och genom att upptäcka och förebygga försämring sparar det lidande för

patienten samtidigt som en plats lämnas ledig för någon annan som behöver den. Sjukhus och avdelningar bör undersöka huruvida lokala riktlinjer följs, och åtgärda eventuella brister i den rutinmässiga övervakningen.

(20)

Cover letter

Dear Editors of Journal of Clinical Nursing,

We are hereby sending you the manuscript of our study “Recognizing deterioration: Nurses’ documentation of vital signs – a systematic literature review”. Research show that abnormal vital parameters can be found up to 48 hours prior to an adverse event. We investigated to what extent vital signs were documented prior to patient deterioration on general wards.

The databases PubMed and CINAHL were searched and nine articles were included in the synthesis. We found that the fraction of cases which had vital signs documented prior to deterioration was diverse, although never reported to be complete. Some studies showed an acceptable fraction of patients who were monitored in the hours prior to an adverse event, but it was seen that the monitoring did not always escalate as the patient got worse. The vital signs most frequently documented were sometimes still missing in a substantial fraction of charts.

These results indicate that there is a lack of adherence in the afferent limb of the rapid

response system, which could potentially prevent activation of the efferent limb and thus be a patient safety risk. It is important to highlight and discuss this trend as the patient care is growing more complex as the population is ageing, and the intensive care beds are scarce.

This work has not been published or submitted for publication elsewhere and we hope that the manuscript is eligible for publication in your journal.

Sincerely,

Frida Collin, MB

School of Medical Sciences Örebro University

Sweden 2020-12-22

(21)

Etisk reflektion

En systematisk litteraturstudie grundar sig på redan existerande forskning. Eftersom arbetet inte inhämtar någon ny information behövs därmed inte etiskt godkännande för denna typ av studie. Av de nio studier som inkluderades i detta arbete hade sju etiskt godkännande, en enbart godkännande från landets ”data protection agency” och en uppgav varken information om eventuellt etiskt godkännande eller gjorde noggranna etiska överväganden. Den

sistnämnda var en retrospektiv observationsstudie, vilket i praktiken därför inte innebar någon risk för påverkan på patientens vård och insamlad data var av sådan natur att det inte går att peka ut enstaka individer.

Ingen av studierna tillfrågade patienterna om samtycke till deltagande. Detta argumenterades för genom att peka på att den retrospektiva observationsstudien till naturen inte kan påverka pågående vård. Att samla in samtycke i efterhand till dess studier skulle dessutom bli svårt ur en praktisk synvinkel, då flera av studierna använde hjärtstopp, allvarlig sjukdom eller död som primära utfall. Alternativet att leta upp överlevande patienter och anhöriga för att inhämta samtycke skulle troligtvis orsaka mer lidande än det skulle göra nytta.

I en systematisk litteraturöversikt är det av betydande vikt att alla relevanta artiklar som kommer upp i sökningen inkluderas, oavsett vad dess data visa. Lika viktigt är det att sedan redovisa och presentera alla data som rör ämnet för att få en sann bild av kunskapsläget. En svårighet med denna studie var att sålla bland de artiklar som hade mycket data om

dokumentation av vitalparametrar, men som gränsade mot att falla bort på exklusionskriterier.