Study reference

Study characteristics

Patient characteristics ²

Intervention (I)

Comparison / control (C) ³

Follow-up

Outcome measures and effect size ⁴

Comments

O’Mara, 2014

ACS model versus On Call

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

USA

Source of funding:

Dr Owens is the CEO and owner of SAMGI, the corporation that contracted with Sutter Medical Center Sacramento to provide the Acute Care Surgery Service. Drs O’Mara, Scherer, and Wisner have all contracted with SAMGI for professional services.

Nothing mentioned about the funding of the study.

Inclusion criteria:

- emergency surgery operations,

Exclusion criteria:

None mentioned

N total at baseline:

Intervention: 428

Control:374

Appendectomy

N total

I:227

C:196

Cholecystectomy

N total

I: 201

C:178

Important prognostic factors²:

Appendectomy

Case mix index

I:1,31 ± 1,42

C:1,29 ± 0,62

NS

Cholecystectomy

Case mix index

I:1,88 ± 1,48

C:1,74 ± 0,77

NS

No other prognostic factors mentioned. It is assumed that groups are comparable at baseline?

Describe intervention:

Acute care surgery

Acute care surgery service was instituted on January

1, 2008. This service consisted of 4 principal general surgeons, all board certified in general surgery, and 2 with additional certification in surgical critical care. These surgeons covered the acute care surgery service in 24-hour rotations, provided all patient care, evaluated all consults, covered the outpatient clinic, and performed all operations and procedures for the service during their hours of coverage.

The surgeons were assisted during their 24-hour rotation by a nurse practitioner or physician assistant.

The acute care surgeons were required to evaluate all emergency and inpatient consults, providing definitive operative or nonoperative care as was defined by management guidelines agreed on by the group.

Describe control:

On Call panel

9 general surgeons constituted the “call panel” and covered call on a rotating basis as part of medical staff requirements, each covering 24 consecutive hours of consults. Surgeons were paid a stipend by the hospital for this coverage. Patients evaluated by a surgeon during this time period were maintained in that surgeon’s practice and cared for by that surgeon and his or her partners. Critical care of these patients was shared by the surgeon and a consulting pulmonary critical care service. Surgeons were required to evaluate all emergency and inpatient consults, with definitive therapy left to the clinical judgment of the surgeon.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Appendectomy

Length of Stay in days (LOS)

I: 1,9 ± 2,1

C: 3,0 ± 3,1

p<0,01

Complications

I: 8/227

C:25/196

p<0,01

Cholecystectomy

Length of Stay in days (LOS)

I: 4,4 ± 5,1

C: 5,3 ± 4,6

p<0,01

Complications

I: 34/201

C:37/178

p<0,02

In this study 4 ACS-periods are described. It was decided to compare the pre-ACS period with ACS-2 because in ACS-1 there might have been adaptation problems.

Fu, 2014

ACS model versus On call

Type of study:

Historically controlled Study

Setting:

Single centre

Country:

Taiwan

Source of funding:

Not mentioned

Inclusion criteria:

- patients with acute appendicitis who underwent appendectomies.

Exclusion criteria:

- Patients younger than 16 years of age;

- Patients who had a septic shock and then underwent exploratory laparotomy.

N total at baseline:

Intervention: 159

Control: 146

Important prognostic factors²:

age ± SD:

I: 43,8 ± 29,0

C: 41,3 ± 16,9

NS

Sex:

I: 55,3 % M

C: 50 % M

NS

Operation during night time

I: 73 %

C: 39%

P<0,001

Groups comparable at baseline.

Describe intervention:

ACS model

In contrast, in-house attending trauma surgeons primarily provided urgent assessment and treatment of nontrauma abdominal surgical emergencies after the implementation of the ACS model (August 2010 to July 2011).

Describe control:

On Call model

the surgical evaluations of patients with abdominal emergencies in the ED were primarily conducted by duty residents of general surgery. Further decision making was supervised by on-call doctors according to the resident’s presentation over the telephone (pre-ACS model).

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Appendectomy

Length of Stay in days (LOS)

I:2,44 ± 1,39

C: 3,83 ± 2,21

Complications

I: 10/159

C: 8/146

NS

Time to operation in min (hours)

I: 300,3 ± 61,7 (5,005 ± 1,03)

C: 719,1 ± 339,0 (11,98 ± 6,65)

p<0,001

Time to operation: from presentation on ED to sending patient to OR

Stupart, 2013

Emergency surgical care model versus On call model

Type of study:

Retrospective cohort

Setting:

Single Centre

Country:

Australia

Source of funding:

Financial support for this study was received from the Department of Health in Victoria. The funding was used to provide sessional payments for the on-site surgeons, and for a project officer to collect the data.

Inclusion criteria:

- Patients undergoing emergency surgery.

Exclusion criteria:

- planned admissions

N total at baseline:

Intervention: 984

Control: 966

Important prognostic factors²:

No prognostic factors mentioned.

Not possible to assess whether groups comparable are at baseline?

Describe intervention:

Emergency surgical care model

implemented a model of emergency surgical care in which daily (Monday–Friday) dedicated emergency general surgery operating sessions were made available, and managed by an on-site consultant surgeon. This surgeon was responsible for deciding which cases could be booked on the list, and their order (according to urgency). The surgeons were also responsible for treating emergency general surgery patients or delegating that care to a more junior colleague at their discretion. The emergency general surgery operating sessions could be made available to other specialties at the discretion of the supervising surgeon according to urgency and the availability of other operating time.

Resources were not available to fund a dedicated emergency surgical unit, and there was strong cultural resistance from some of the involved surgeons to the concept of routine handover of patients (a potential problem of implementation identified by Parasyn et al.3).

Describe control:

On Call model

The general surgery emergency on call service is delivered by 15

surgeons, most of whom have subspecialist private practices.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Median Length of Stay in days (LOS)

I: 3,0 (3,0 - 3,0)

C: 3,0 (3,0 - 4,0)

P=1,3

Median Time to operation in hours

I: 18 (17-19)

C: 19 (18-21)

P=0,033

Time to operation: from presentation on ED to sending patient to OR

Note: The times from ED to theatre, from booking to theatre and hospital stay, were not normally distributed, so the figures are presented as median (95% confidence interval for median) and were compared using the Mann–Whitney U-test.

Poh, 2013

ASU model versus On call model

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

Australia

Source of funding:

Not mentioned

Inclusion criteria:

Patiens:

- Aged 16-99 years;

- underwent emergency appendicectomies.

Exclusion criteria:

- elective surgery.

N total at baseline:

Intervention: 283

Control: 256

Important prognostic factors²:

age ± SD:

I: 31,88 ± 0,89

C: 31,52 ± 0,85

Sex:

I: 47 % M

C: 44,5 % M

NS

Groups comparable at baseline.

Describe intervention:

ASU model

The key aspects of the ASU were as follows: Firstly, the availability of an on-site general surgeon from 08.00-17.00 hours every day, who was then exclusively on-call after-hours. This promoted consultant-led patient care. Secondly, the availability of a dedicated ASU theatre every afternoon (13.00–17.00 hours), except Sunday, to cater for the steady flow of expected ‘emergency’ patients. The general emergency operating list remained available to the unit and was often used for ASU patients in the morning since the on-site consultant surgeon was often available during these hours. Thirdly, there was a dedicated unit structure consisting of one consultant, two registrars, one resident and two interns. Consultants rotated through a 24-hour shift, beginning at 19.00 hours daily. Continuity of care was preserved at a registrar and resident level. The management of all patients on the ASU was handed over from one surgeon to the next, unless the patients had been operated on and were expected to require a post-operative stay of greater than 48 h, in which case they were handed over to the appropriate specialist unit. Finally, in the ASU model, surgeons were remunerated at an hourly rate with a fixed on-call moiety. In addition to recognizing the surgeons’ commitment to the ASU, this also eliminated the financial incentive for after-hours surgery

Describe control:

On Call model

the on-call consultant would handle the emergency case load on top of his routine commitments. The consultants were all visiting medical officers (except one full-time consultant appointed in January 2011) with responsibilities at a number of other hospitals. Typically, emergency patients remained under the care of the surgeon on-call when they were admitted. There was an ‘acute surgical registrar’ who assessed the acute surgical patients, but ongoing management at the registrar level was provided by the surgical registrar on the unit to which the consultant on-call belonged. Patients who required surgery were booked on the general emergency list and had to compete with patients from other surgical services. As the surgeons were often committed elsewhere during office hours and there was no guarantee of theatre availability during daylight hours, patients who required surgery were often operated on after-hours despite the fact that there was no pressing need for an operation at this time

Length of follow-up:

1 month

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Intervention:

N=19 (6,7 %)

Reasons (data on time to operation incomplete)

Control:

N=13 (5,1%)

Reasons (data on time to operation incomplete)

No reason to believe that the incomplete outcome data introduced bias.

Outcome measures and effect size:

Median Time to operation in min IQR

I: 1129 (617 – 1468) = 18,8 hours

C: 1080 (619 – 1467) = 18 hours

P=0,963

Median Length of Stay in days IQR (LOS)

I: 2 (2 – 3)

C: 2 (2 – 3)

P=0,16

Complications (surgical site infections)

I: 8/283

C: 7/256

Lim 2013

ACES model versus On Call model

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

Canada

Source of funding:

Not mentioned

Inclusion criteria:

Exclusion criteria:

- Patients treated nonoperatively;

- elective surgery;

- primary admission other than biliary tract disease.

N total at baseline:

Intervention: 172

Control: 72

Important prognostic factors²:

age:

I: 51,2

C:49,7

NS

Sex:

I: 44,8% M

C: 51,4% M

No significant differences in numerous co-morbidities.

Groups comparable at baseline.

Describe intervention:

ACES model

The Acute Care Emergency Surgery (ACES) service is comprised of nine dedicated surgeons who rotate weekly, as well as a senior resident, two junior residents, and two hospitalists. For a particular week, the designated ACES surgeon and team is on call daily from 07h00 to 19h00 (12 h). At 19h00, a second surgeon and resident team (which varies nightly) takes over call duties until 07h00 the following morning, at which time care is handed back to the designated ACES team. Handovers therefore occur every morning at 07h00 and every evening at 19h00 between the daytime ACES team and the variable nighttime ACES team. The ACES service is unique because of funding that allows for a dedicated emergency general surgery daytime operating room that operates during weekday daytime hours from 07:30 to 15:30.

Describe control:

On Call model

Emergency cases are added to the on-callsurgeon’s elective schedule.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Complications

I: 39/172

C: 25/72

Time to operation in hours

I: 21,5 ± 14,2

C:31,1 ± 21,4

P<0,001

Length of Stay in days (LOS)

I: 6,0 ± 5,4

C: 6,0 ± 4,5

NS

Faryniuk, 2013

ACSS mode versus Traditional model

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

Canada

Source of funding:

Not mentioned

Inclusion criteria:

- Patients with acute appendicitis, acute cholecystitis and small bowel obstruction;

Exclusion criteria:

Not mentioned.

N total at baseline:

Intervention: 137

Control: 67

Important prognostic factors²:

For example

age ± SD:

I: 49,7

C: 50,8

NS

Sex:

I: 51 % M

C: 57 % M

Groups comparable at baseline.

Describe intervention:

ACCS model

The ACSS model described in this study has a dedicated surgeon committed for a full week, Monday to Sunday, 8 am to 4 pm, focused on the efficient care of the emergent nontrauma surgical patient. Surgeons are thereby allowed to focus completely on their elective practices when not on the ACSS.

Describe control:

Traditional model

traditional model has relied on an elective subspecialized service to take on the emergent surgical patient load. Tight scheduling of operative, clinic, research and academic time has made care for emergency surgical patients in this setting exceedingly difficult and often illtimed. As a result, emergency procedures have often been done at the end of busy elective slates or through the night, with a negative impact on the surgeon’s daytime schedule.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Mean Length of Stay in min

I: 5677,6 = 3,9 days

C: 5803,8 = 4,02

P=0,91

Mean Time to operation in min

I: 453,4 = 7,55

C: 850,5 = 14,17

P=0,01

Three periods are described in the study. The pre-ACCS period, the newly formed ACSS period and the established ACCS period. It was decided to compare the pre-ACCS period with the established ACCS period.

Cubas, 2012

ACS model versus On Call model

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

USA

Source of funding:

Not mentioned

Inclusion criteria:

- Aged ≥15 years;

- acute abdomen of appendicular or cholecystic origin.

- underwent urgent operation.

Exclusion criteria:

- Elective procedures;

- procedures as part of another major operation;

- patients with associated complications or who were treated conservatively

N total at baseline:

Appendectomy

Intervention:93

Control:82

Cholecystectomy

Intervention: 62

Control:51

Important prognostic factors²:

Appendectomy

age ± SD:

I: 34 ± 17

C: 35 ± 17

NS

Sex:

I: 55 % M

C: 43 % M

NS

Cholecystectomy

age ± SD:

I: 40 ± 19

C: 38 ± 17

NS

Sex:

I: 21 % M

C: 16 % M

NS

Groups comparable at baseline.

Describe intervention:

Acute care surgery model (ACS)

The acute care surgery service model, designed to combine trauma and emergency general surgery divisions into one 12-hour in-house service.

Describe control:

On Call model

Traditionally, nontrauma patients needing emergency surgical care had to wait for the general surgeon on call, who was usually preoccupied with other activities, such as office consults or elective surgery, which often occur outside of the hospital.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Appendectomy

Complications

I: 9/93

C: 16/82

P=0,06

Mean Time to operation in h

I: 11

C: 16,4

P=0,006

Mean Length of Stay in days

I: 1,8

C: 2,8

P=0,002

Cholecystectomy

Complications

I: 5/62

C:13/51

P=0,01

Mean Time to operation

I: 34

C:60

P=0,002

Mean Length of Stay in days

I: 3,3

C:5,2

P=0,03

Lau, 2011

ACS model versus Traditional model

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

USA

Source of funding:

Not mentioned

Inclusion criteria:

- aged ≥18 years of age;

- Acute cholecystitis or early cholecystectomy.

Exclusion criteria:

- Choledocholithiasis, biliary colic or cholelithiasis without cholecystis;

- treated nonoperatively;

N total at baseline:

Intervention: 71

Control: 81

Important prognostic factors²:

age ± SD:

I: 45,7

C: 46,7

NS

BMI kg/m²

I: 31,1

C: 29,6

Sex:

I: 28% M

C: 30% M

Groups comparable at baseline.

Describe intervention:

ACS model

ACS team was composed of an attending surgeon who was free from all other daytime clinical duties and one or two residents. No OR was exclusively available for the on-call team.

Describe control:

Traditional model

The responsible attending surgeon had other scheduled clinical duties such as outpatient clinic or elective operations. No OR was exclusively available for the on-call team.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Time to operation in h

I: 24,6

C: 35,0

P=0,0276

Length of Stay in days (LOS)

I: 2,63

C: 3,23

P=0,11

Complications

I: 5/71

C: 15/81

Lehane, 2010

ACS model versus On Call model

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

Australia

Source of funding:

There has been no research grant.

Inclusion criteria:

- Admitted for acute cholecystitis.

Exclusion criteria:

Not mentioned.

N total at baseline:

Intervention: 115

Control: 87

Important prognostic factors²:

For example

age ± SD:

I: 47 ± 17,0

C: 50 ± 19,5

Sex:

I: 30 % M

C: 38 % M

Groups comparable at baseline.

Describe intervention:

ACS model

An ACS model was introduced in September 2005. The on-call roster was altered, dividing the week into three periods: Monday (08.00 hours)– Wednesday (12.30 hours), Wednesday (12.30 hours)–Friday (18.00 hours) and Friday (18.00 hours)–Monday (08.00 hours). The acute care surgeons rotated through these periods. During their on call period the surgeon was available on-site from 08.00 until 18.00 hours and had no other commitments. On weekdays between 18.00 and 08.00 hours and on weekends the consultant was available but not on site unless required. They also had a dedicated surgical registrar and resident team to manage the ACS patients, ensuring continuity of care. The consultant had full control of the hospitals emergency theatre. This theatre is shared with other specialties including plastics, neurosurgery, orthopaedics, otolaryngology, vascular and renal transplant, ophthalmology and paediatric surgery. All obstetric surgery is undertaken at a separate facility. The ACS surgeon was responsible for the order of cases. This version of an ACS model was created with a view that early consultant presence should fast track appropriate cases and maximize theatre allocation. At the end of the on-call period the consultants’ team would then formally take over all patients whom had received ‘definitive care’.

Describe control:

On Call model

It involves the on-call surgical registrar assessing the patient and relaying the salient information to the consultant, who may or may not be present on site. The decision is then made as to whether further investigation or definitive surgical intervention is required.

Prince of Wales Hospital had a general surgical on-call system

based on the traditional consultant-attending model.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Complications

I: 10/115

C:15/87

Median Length of Stay in days

I: 4

C: 6

P=0,001

Gandy, 2010

ACS model versus On Call Model

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

Australia

Source of funding:

Not mentioned

Inclusion criteria:

- Underwent appendectomy;

Exclusion criteria:

- Elective surgery;

N total at baseline:

Intervention: 226

Control: 176

Important prognostic factors²:

age:

I: 32,8

C: 33,6

NS

Sex:

I: 54% M

C: 47,2% M

NS

Groups comparable at baseline.

Describe intervention:

ACS Model

Separating the elective and emergency workloads and placing control of the emergency theatre under the acute care general surgeon. Control of the emergency operating theatre allowed for planning of the semiurgent caseload. The duty acute care surgeon remained on site during the working day, with no other commitments other than the ACS service. Patients whose surgical care had not been completed by the end of the duty period were handed on to the next acute care duty surgeon for definitive care. The introduction of the ACS service at our institution coincided with the formalised referral of surgical patients from a second emergency department within the area network. There were no changes in the clinical guidelines for patient care.

Describe control:

On Call model

Traditional on call model.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Complications

I: 21/226

C: 30/176

P=0.02

Median Length of Stay in days (LOS)

I: 3 (1-13)

C:2 (1-17)

P=0,92

Median Time to operation in hours

I: 9:36 = 576 minutes = 9,6 h

C: 10:27 = 627 minutes = 10,45 h

P=0,29

Britt, 2010

ACS model versus On Call model

Type of study:

Historically Controlled Study

Setting:

Single Centre

Country:

USA

Source of funding:

Not mentioned

Inclusion criteria:

- patients with biliary disease.

Exclusion criteria:

- Elective cases.

N total at baseline:

Intervention: 132

Control:54

Important prognostic factors²:

age:

I: 44,4

C:50,2

P=0,05

Sex:

I: 23,5 % M

C: 37% M

Groups comparable at baseline.

Describe intervention:

ACS model

In July 2007 wij instituted an ACS service staffed by 6 board-certified critical care surgeons to provide all inpatient and emergency room consults. The ACS service is staffed by a PGY-4 and PGY-2 resident team.

Describe control:

On Call model

surgical evaluations in the emergency department as well as inpatient consultations were done by a rotating group of academic surgeons based on a call schedule, with the patients followed by the professorial

academic surgical service.

Length of follow-up:

Not mentioned

Loss-to-follow-up:

Not mentioned

Incomplete outcome data:

Not mentioned

Outcome measures and effect size:

Time to operation in m

I: 59,9 ± 65,5

C: 68,7 ± 84,9

P=0,045

Length of Stay in days

I: 5,5 ± 5,6

C: 6,7 ± 8,8

P=0,027

Complications

I: 11/132

C:10/54

P=0,06

Study reference

(first author, year of publication)

Bias due to a non-representative or ill-defined sample of patients?¹

(unlikely/likely/unclear)

Bias due to insufficiently long, or incomplete follow-up, or differences in follow-up between treatment groups?²

(unlikely/likely/unclear)

Bias due to ill-defined or inadequately measured outcome ?³

(unlikely/likely/unclear)

Bias due to inadequate adjustment for all important prognostic factors?⁴

(unlikely/likely/unclear)

O’Mara, 2014

Unlikely

Unlikely

Unlikely

Unclear

Fu, 2014

Unlikely

Unlikely

Unclear

Unclear

Stupart, 2013

Unclear

Unlikely

Unclear

Unclear

Poh, 2013

Unlikely

Unlikely

Unlikely

Unclear

Lim, 2013

Unlikely

Unlikely

Unlikely

Unclear

Faryniuk, 2013

Unlikely

Unlikely

Unlikely

Unclear

Cubas, 2012

Unlikely

Unlikely

Unlikely

Unclear

Lau, 2011

Unlikely

Unlikely

Unlikely

Unclear

Lehane, 2010

Unlikely

Unlikely

Unlikely

Unclear

Gandy, 2010

Unlikely

Unlikely

Unlikely

Unclear

Britt, 2010

Unlikely

Unlikely

Unlikely

Unclear