| | |

Received: 25 January 2018    Revised: 7 April 2018    Accepted: 8 June 2018    First published online: 5 July 2018

DOI: 10.1002/ijgo.12565

CLINICAL ARTICLE
Obstetrics

Long-­term outcomes of the Stop Traumatic OASI Morbidity

Project (STOMP)

Maya Basu* | Dot Smith

Department of Obstetrics and Gynaecology,

Medway NHS Foundation Trust,
Gillingham, UK
*Correspondence
Maya Basu, Consultant Obstetrician and
Gynaecologist, Medway NHS Foundation
Trust, Windmill Road, Gillingham, Kent, UK.
Email: mayabasu@aol.com
Abstract
Objective: To evaluate long-­term sustainability of the Stop Traumatic OASI Morbidity
Project (STOMP) in reducing the incidence of obstetric anal sphincter injury (OASI).
Methods: A prospective observational study of women undergoing vaginal delivery at
a UK district general hospital between September 1, 2014, and February 28, 2017.
The principles of STOMP involve encouraging upright positioning, verbal coaching to
avoid expulsive pushing and to slow down delivery, and tactile support to the vertex
to judge speed and slow down delivery. After a training period, STOMP was imple-
mented for all vaginal deliveries. Clinical and demographic data on women affected by
OASI were collected across a 30-­month period. The primary outcome measure was
the incidence of OASI.
Results: There were 8782 vaginal deliveries during the 30-­month period after imple-
mentation of STOMP. There was a significant decrease in the mean incidence of OASI
relative to the 9 months before implementation (P<0.001). There was a significant
decrease in the incidence of OASI for both spontaneous vaginal and instrumental
deliveries (both P<0.05). There was no change in the frequency of episiotomy.
Conclusions: Implementation of STOMP led to a significant decrease in OASI, confirm-
ing the sustainability of this approach to improve outcomes.

KEYWORDS
Anal incontinence; Maternal childbirth injury; Obstetric anal sphincter injuries; Perineal trauma;
Prevention; Quality improvement

1 |  INTRODUCTION More recently, a prospective interventional program of changes in

Worldwide, there has been a significant rise in the incidence of obstet-
ric anal sphincter injury (OASI) over the past two decades. Women
with OASI are more likely to be affected by anal incontinence, sexual
dysfunction, and impairment of overall quality of life.1–3
Historically, there has been much focus on the diagnosis and
management of OASI, which has led to improvements in outcomes
for women.4 Nevertheless, primary prevention of this adverse event
of pregnancy should be the overarching goal. Previous evaluations
of strategies to reduce the risk of severe perineal trauma, including
“hands on” and episiotomy, have yielded conflicting results.5–7
delivery practices at a single unit in Norway reported a reduction in
OASI incidence from 4.03% to 1.17%.8 The program was subsequently
implemented in a multicenter context, and a similar decrease in the
rate of OASI was seen.9
In the United Kingdom, the Stop Traumatic OASI Morbidity
Project (STOMP), which was developed to reduce the incidence of
OASI, led to a reduction in OASI rate from 4.7% to 2.2% at 1 year
after implementation.10 However, those short-­term results are not
informative in terms of the “real-­world” sustainability of implemen-
tation of the approach after the initial launch phase. The aim of the
present study was therefore to evaluate long-­term changes in the

Int J Gynecol Obstet 2018; 142: 295–299 © 2018 International Federation of |  295
wileyonlinelibrary.com/journal/ijgo  
Gynecology and Obstetrics
 |
296       Basu and Smith
incidence of OASI in the 30 months following implementation of
STOMP in August 2014.
2 |  MATERIALS AND METHODS
In the present prospective observational study, delivery outcomes
were reviewed for the 30-­month period following full implementa-
tion of the STOMP quality improvement bundle at Medway Maritime
Hospital, Gillingham, UK, a large district general hospital, on September
1, 2014. STOMP was registered as a quality improvement project with
the Audit and Quality Improvement Department of Medway NHS
Foundation Trust. Under current UK guidance, formal ethical approval
was not required because the study was a quality improvement project
and not a clinical trial. For the same reason, written informed consent
was not required; however, all women were told about STOMP prena-
tally, and all women’s delivery preferences were adhered to.
All women undergoing vaginal delivery were eligible for inclusion
in the study. Cesarean delivery was the only exclusion criterion. Full
details of the background, design and implementation of STOMP have
been published elsewhere.10 The STOMP intervention has three key
clinical practice components: position, coach, and speed.
The position component includes: avoiding the semi-­recumbent
and lithotomy positions (not applicable to assisted vaginal deliveries);
and encouraging upright positioning. Visualization of perineum at
delivery was considered unnecessary.
The coach component includes: effective communication and
coaching during the active second stage to promote more controlled
and less expulsive maternal efforts. Avoidance of directed pushing and
encouragement to follow perceived urge to push. Verbal encourage-
ment to slow down expulsive efforts at crowning.
The speed component includes: simple tactile control (flat of hand
placed on head) to gauge speed and allow for coaching to slow down
delivery of the head (not manual manipulation of vertex/perineal ring,
pinching of fourchette, or Ritgen maneuver). During assisted vaginal
delivery, use of slow controlled delivery with minimal or no traction
as head crowns. An aim of spontaneous delivery of the shoulders or
minimal traction for delivery of shoulders was included.
In addition, all staff received mandatory training on the recognition
of OASI. Third and fourth degree tears were classified and repaired in
line with national guidance.
The STOMP bundle also encompasses a monthly audit of
patient records. Owing to a rise in OASI incidence 6 months after
T A B L E   1   Incidence of OASI per quarter following full implementation of STOMP.
Quarterly period from September 1, 2014, to February 28, 2017
Outcome 1 2 3 4
No. of vaginal deliveries 892 862 885 881
No. of OASI cases 19 20 22 21
OASI rate, % 2.1 2.3 2.5
Abbreviations: OASI, obstetric anal sphincter injury; STOMP, Stop Traumatic OASI Morbidity Project.
implementation, two further measures were added to the bun-
dle: STOMP stickers with a checklist of the three components for
documentation purposes, and peer review of deliveries, whereby
compliance with the three STOMP components was evaluated and
documented by a second midwife or doctor at delivery rather than
by the accoucheur themselves.
For the present study, the primary outcome measure was the
incidence of OASI (expressed as a percentage of total vaginal deliv-
eries), which was collected and recorded on a monthly basis. The
incidence of OASI was ratified on a 3-­monthly basis. Each OASI was
individually evaluated, and data were collected on relevant clinical and
demographic factors. Audit data were checked against the maternity
database to ensure accuracy.
To calculate OASI rates, data on the mode of delivery and parity for
all women who delivered between September 1, 2014, and February
28, 2017, were obtained from the maternity database. As a pre-­STOMP
reference population, clinical and demographic data from women who
delivered between October 1, 2013, and June 30, 2014, before the
STOMP training month were used for comparative purposes.
Statistical calculations were performed with SPSS version 20 (IBM,
Armonk, NY, USA). The χ2 test (with Yates correction) was used to
compare categoric data, and the Student t test or Mann-­Whitney U
test was used to compare continuous data. P<0.05 was considered
statistically significant.
3 | RESULTS
In the 30 months following implementation of STOMP, there were
8782 vaginal deliveries at the study hospital and 196 recorded cases
of OASI.
The baseline incidence of OASI in the 9 months before full STOMP
implementation (i.e., excluding the pre-­implementation awareness and
education training month) was 4.6% (138/3007 vaginal deliveries).
The mean incidence of OASI per quarter since implementation is given
in Table 1. At month 30, the year-­to-­date mean incidence of OASI was
1.8% (66/3690). As compared with the baseline incidence of 4.6%,
this represents a significant decrease in OASI incidence at 30 months
after the implementation of STOMP (P<0.001).
During the evaluation period, 129/196 (65.8%) of the cases of
OASI occurred after spontaneous vaginal delivery; of these, 53/196
(27.0%) and 14/196 (7.1%) occurred after forceps and vacuum
­deliveries, respectively.
5 6 7 8 9 10
915 868 931 825 884 839
23 18 19 22 13 19
2.4 2.5 2.1 2.0 2.7 1.5 2.3
Basu and Smith |
      297

Among the 196 cases of OASI, there were 113 (57.7%) grade T A B L E   3   Incidence of OASI stratified by mode of delivery and
3a, 52 (26.6%) grade 3b, 18 (9.2%) grade 3c, and 13 (6.5%) grade 4 parity in pre-­ and post-­STOMP cohorts.a
tears. There was no difference in the proportion of each type of tear
Pre-­STOMP STOMP cohort
between the data at 12 months and those at 30 months of implemen- cohort (n=3007)b (n=8782)c
tation (data not shown).
Factor Patients OASI Patients OASI P value
Clinical and demographic data for all women who delivered during
the 9-­month period pre-­STOMP and the 30 months post-­STOMP Mode of delivery

implementation are given in Table 2. The parity and mode of delivery Vaginal 2671 74 (2.8) 7699 129 (1.7) <0.001
of all eligible women (i.e., excluding those who delivered by cesar- Forceps 214 33 (15.4) 668 53 (7.9) 0.002
ean), together with the overall incidence of OASI by parity and mode Vacuum 122 12 (9.8) 415 14 (3.4) 0.007
of delivery pre-­ and post-­STOMP, are given in Table 3. The overall Parity
year-­to-­date rate of episiotomy at month 30 after implementation 0 1364 93 (6.8) 3502 133 (3.2) <0.001
(384/3690; 10.4%) was not significantly different from that at baseline ≥1 1643 28 (1.7) 5280 83 (1.6) 0.736
(372/3656; 10.2%) (data not shown).
Abbreviation: OASI, obstetric anal sphincter injury; STOMP, Stop
Traumatic OASI Morbidity Project.
a
Values are given as number or number (percentage), unless indicated
4 | DISCUSSION otherwise.
b
From October 1, 2013, to June 30, 2014.
c
From September 1, 2014, to February 28, 2017.
The present study reported delivery outcomes for the 30-­month
period following implementation of STOMP. These longer-­term data
indicated that the STOMP bundle has continued to lead to a reduction address the conduct of the second stage in its entirety, but also to gen-
in the incidence of OASI (to a yearly mean of 1.8%), even though the erate enthusiasm and engagement for the project via education and
initial implementation phase has passed. In addition, there has been a publicity. In other words, OASI prevention could be better achieved by
significant decrease in the rate of OASI across all modes of delivery. multifaceted campaigns that focus on how the second stage of labor
The study had potential limitations. First, because STOMP is a is conducted, as well as by awareness, training, and engagement of
package, it is not possible to evaluate the impact of individual com- healthcare professionals.
ponents on OASI rates. The factors resulting in OASI are complex; Second, the present evaluation of STOMP was carried out at a
therefore, a single intervention may not have a significant effect on its single center, meaning that its generalizability could be limited. To
incidence. Indeed, randomized controlled trials have failed to consis- the best of our knowledge, however, the present study is the first
tently identify single interventions that can prevent OASI, highlighting to report long-­term outcomes from a multi-­intervention quality
its complexity. By contrast, the central idea underlying the STOMP improvement bundle implemented outside Scandinavia. The strength
project is that, in order to effect change, it is necessary not only to of the STOMP bundle is the simplicity of the intervention itself, as
discussed below.
T A B L E   2   Demographic and clinical characteristics pre-­ and Other interventions have involved “traditional” internal maneu-
post-­STOMP implementation.a vers to facilitate flexion and delivery of the vertex, as well as manual
Pre STOMP Post STOMP perineal protection.8 The aim of STOMP was to develop an alternative
Characteristic (n=3667)b (n=12 414)c P value intervention without manual perineal protection, but with previously

Age, y 32 ± 6 31 ± 6 0.762 described mandated delivery positions.8 Thus, the only mandated
contact component of STOMP involves tactile support to the vertex
BMI 26.6 ± 7.8 26.8 ± 6.1 0.448
with a single hand. Attempts to reproduce the Norwegian interven-
Parity 1.8 2.0 0.320
tion in other settings have yielded conflicting results,11,12 and a review
Delivery weight, g 3327 ± 649 3343 ± 629 0.296
of those data concluded that more information is required about
Mode of delivery
effects associated with the intervention, such as reducing mater-
Instrumental 315 (8.6) 1117 (9.0) 0.531
nal choice with reference to delivery position and increasing rates
Forceps 165 (4.5) 621 (5.0) 0.231 of episiotomy.13
Cesarean 935 (25.5) 3476 (28.0) 0.056 The data on manual perineal protection and tactile support during
Induction of labor 862 (23.5) 3712 (29.9) 0.001 delivery as a single intervention are conflicting. A recent systematic
Abbreviations: BMI, body mass index (calculated as weight in kilograms review and meta-­analysis of the use of manual perineal protection
divided by the square of height in meters); OASI, obstetric anal sphincter alone (i.e., when not nested within an improvement bundle) in the pre-
injury; STOMP, Stop Traumatic OASI Morbidity Project. vention of OASI failed to show a protective effect14; however, evalu-
a
Values are given as mean ± SD, mean, or number (percentage), unless indi-
ation of the data was hampered by heterogeneity in the definition of
cated otherwise.
b
From October 1, 2013, to June 30, 2014. the intervention used. As mentioned above, the causes of OASI are
c
From September 1, 2014, to February 28, 2017. highly complex; thus, comparative studies evaluating a single factor
 |
298       Basu and Smith
such as manual perineal protection may not show an effect because
they do not take into account other factors that may modulate the
risk of OASI.
Notably, the reduction in OASI rate observed in the study cohort
was not associated with a change in episiotomy rates, as described for
9,15
other interventions. Previous studies have indicated that episiotomy
may be protective against OASI,16,17 and that the angle of the episiot-
omy cut affects outcomes.18,19 The OASI care bundle, which has been
launched by the UK Royal College of Obstetricians and Gynaecologists,
includes optimization of the angle of episiotomy as one of its compo-
nents. Bespoke scissors, which allow for standardization of the angle,
have been advocated by some20; however, such specialist pieces of
equipment will have cost implications, and there is evidence that their
21
introduction can lead to an increase in intervention. Furthermore,
although episiotomy may be potentially protective against OASI in
some settings, other data indicate that it is associated with an increased
risk of OASI.7 Episiotomy also has associated morbidity including hem-
22,23
orrhage, pain, and wound infection or breakdown. The present
results are consistent with another report indicating that modifica-
tion of second stage practices does not need to be associated with an
increase in episiotomy in order to decrease OASI rates.24
Both forceps and vacuum delivery are known to be significant risk
factors for OASI.25 In the present study, the OASI rate decreased sig-
nificantly for both spontaneous deliveries and instrumental deliveries,
which demonstrates that the STOMP principles applied at delivery
can mitigate the risk of OASI even for deliveries that carry a higher
overall risk. The known increase in risk with instrumental deliveries
means that strategies to reduce OASI incidence for these deliveries
have received less attention. Although lithotomy positioning cannot
be avoided for an instrumental delivery, the principles of coaching
and slow delivery of the head, and careful delivery of the shoulders
can be applied, and were associated with a clear decrease in the inci-
dence of OASI among women in this higher risk group. Other studies
have focused on the use of episiotomy during instrumental delivery to
reduce rates of OASI; however, healthcare staff in the present study
were not instructed to change their practice with regard to episiotomy
and the episiotomy rates were unchanged.
Nulliparity is also a known risk factor for OASI, and STOMP was
associated with a measurable decrease in OASI incidence in this
group. The present results support the concept that the incidence of
OASI can be reduced for higher risk deliveries (e.g., nulliparity or for-
ceps delivery), which is relevant because this group has the greatest
scope for improvement.
The present data indicate that the positive effects of STOMP
on OASI rates were sustained in the longer term. Whereas short-­
term effects can be attributed to some degree to the initial high
enthusiasm and engagement of staff, the longer-­term data demon-
strate that the principles of STOMP can be embedded into prac-
tice and unit culture, leading to ongoing change. STOMP is a simple
and low-­cost intervention with the potential for transfer to other
settings. Its implementation into individual units relies heavily on
ensuring that staff engage with the project with a willingness to
change clinical practice; this is the main challenge of such a project.
STOMP was promoted both by an educational program and regu-
lar electronic communications to explain to staff why reductions in
OASI rates is an important issue, and by an accompanying publicity
campaign to keep the program as dynamic and exciting as possible.
Implementation of STOMP was not associated with an increase in
the rate of episiotomy, thereby avoiding a rise in morbidity and costs
associated with that intervention.
Obstetric anal sphincter injury remains an important key per-
formance indicator for maternity services. Although improvements
in training for assessment and repair of OASI have undoubtedly led
to better outcomes for women, OASI remains an important cause of
impaired quality of life for affected women. Primary prevention is
of upmost importance in reducing the burden of physical and psy-
chologic morbidity in the obstetric population, as well as the asso-
ciated pressures on gynecologic and colorectal services caused by
later adverse events. Future work will focus on evaluating whether
quality improvement bundles such as STOMP are transferable to
other settings.
AU T HO R CO NT R I B U T I O NS
MB contributed to the conception and design of the study, data col-
lection and analysis, and writing the manuscript. DS contributed to
the conception and design of the study, and writing the manuscript.
CO NFL I C TS O F I NT ER ES T
The authors have no conflicts of interest.
REFERENCES
1. Keighley M, Perston Y, Bradshaw E, Hayes J, Keighley D, Webb S. The
social, psychological, emotional morbidity and adjustment techniques
for women with anal incontinence following Obstetric Anal Sphincter
Injury: Use of a word picture to identify a hidden syndrome. BMC
Pregnancy Childbirth. 2016;16:275.
2. Halle T, Salvesen K, Volloyhaug I. Obstetric anal sphincter injury and
incontinence 15-­23 years after vaginal delivery. Acta Obstet Gynecol
Scand. 2016;95:941–947.
3. Stedenfeldt M, Pirhonen J, Blix E, Wilsgaard T, Vonen B, Oian P. Anal
incontinence, urinary incontinence and sexual problems in primipa-
rous women-­ a comparison between women with episiotomy only
and women with episiotomy and obstetric anal sphincter injury. BMC
Womens Health. 2014;14:157.
4. Andrews V, Thakar R, Sultan A. Outcome of obstetric anal sphincter
injuries (OASIS)-­ role of structured management. Int Urogynecol J
Pelvic Floor Dysfunct. 2009;20:973–978.
5. Aasheim V, Nilsen A, Lukasse M, Reinar L. Perineal techniques during
the second stage of labour for reducing perineal trauma. Cochrane
Database Syst Rev. 2011;(12):CD00672.
6. Twidale E, Cornell K, Liitzow N, Hotchin A. Obstetric anal sphincter
risk factors and the role of mediolateral episiotomy. Aus N Z J Obstet
Gynecol. 2013;53:17–20.
7. Andrews V, Sultan A, Thakar R, Jones P. Risk factors for obstetric anal
sphincter injury: A prospective study. Birth. 2006;33:117–122.
8. Laine K, Pirhonen T, Rolland R, Pirhonen J. Decreasing the inci-
dence of anal sphincter tears during delivery. Obstet Gynecol.
2008;111:1053–1057.
Basu and Smith
9. Hals E, Oian P, Pirhonen T, et  al. A multicenter interventional pro-
gram to reduce the incidence of anal sphincter tears. Obstet Gynecol.
2010;116:901–908.
10. Basu M, Smith D, Edwards R. STOMP Project Team. Can the incidence
of obstetric anal sphincter injury be reduced? The STOMP experience.
Eur J Obstet Gynecol Reprod Biol. 2016;202:55–59.
11. Leenskjold S, Ho L, Pirhonen J. Manual protection of the perineum
reduces the risk of obstetric anal sphincter ruptures. Dan Med J.
2015;62:Pii: A5075.
12. Skriver-Moller A, Madsen M, Poulsen M, Overgaard C. Do we know
enough? A quality assessment of the Finnish intervention to pre-
vent obstetric anal sphincter injuries. J Matern Fetal Neonatal Med.
2016;29:3461–3466.
13. Poulsen M, Madsen M, Skriver-Moller A, Overgaard C. Does the
Finnish intervention prevent obstetric anal sphincter injuries? A sys-
tematic review of the literature. BMJ Open. 2015;5:e008346.
14. Bulchandani S, Watts E, Sucharitha A, Yates D, Ismail K. Manual peri-
neal support at the time of childbirth: A systematic review and meta-­
analysis. BJOG. 2015;122:1157–1165.
15. Fretheim A, Odgaard-Jensen J, Rottingen J, Reinar L, Vangen S, Tanbo
T. The impact of an intervention programme employing a hands-­on
technique to reduce the incidence of anal sphincter tears: Interrupted
time-­series reanalysis. BMJ Open. 2013;3:e003355.
16. Verghese T, Champaneria R, Kapoor D, Latthe P. Obstetric anal sphinc-
ter injuries after episiotomy: Systematic review and meta-­analysis. Int
Urogynecol J. 2016;27:1459–1467.
17. Lund N, Persson L, Jango H, Gommesen D, Westergaard H. Episiotomy
in vacuum-­assisted delivery affects the risk of obstetric anal sphincter
|
      299
injury: A systematic review and meta-­analysis. Eur J Obstet Gynecol
Reprod Biol. 2016;207:193–199.
18. Eogan M, Daly L, O’Connell P, O’Herlihy C. Does the angle of episi-
otomy affect the incidence of obstetric anal sphincter injury? BJOG.
2006;113:190–194.
19. Stedenfeldt M, Pirhonen J, Blix E, Wilsgaard T, Vonen B, Oian P.
Episiotomy characteristics and risks for obstetric anal sphincter inju-
ries: A case control study. BJOG. 2012;119:724–730.
20. Freeman R, Hollands H, Barron L, Kapoor D. Cutting a mediolateral
episiotomy at the correct angle: Evaluation of a new device, the
Episcissors-­60. Med Devices (Aukl). 2014;7:23–28.
21. van Roon Y, Kirwin C, Rahman N, et al. Comparison of obstetric anal
sphincter injuries in nulliparous women before and after introduction
of the EPI-­SCISSORS-­60 at two hospitals in the United Kingdom. Int J
Womens Health. 2015;7:949–955.
22. Sagi-Dain L, Sagi S. Morbidity associated with episiotomy in vac-
uum delivery: A systematic review and meta-­analysis. BJOG.
2015;122:1073–1081.
23. Macleod M, Strachan B, Bahl R, et al. A prospective cohort study of
maternal and neonatal morbidity in relation to use of episiotomy at
operative vaginal delivery. BJOG. 2008;115:1688–1694.
24. Rasmussen OB, Yding A, Anh ØJ, Sander Andersen C, Boris J. Reducing
the incidence of Obstetric Sphincter Injuries using a hands-­on tech-
nique: An interventional quality improvement project. BMJ Qual
Improv Rep. 2016;5:pii: u217936.w7106.
25. Hehir M, O’Connor H, Higgins S, et al. Obstetric anal sphincter injury,
risk factors and method of delivery-­ an 8-­year analysis across two ter-
tiary referral centres. J Matern Fetal Neonatal Med. 2013;26:1514–1516.