A Nationwide Flash-Mob Study for Suspected Acute Coronary Syndrome

Angel M.R. Schols; Robert T. A. Willemsen; Tobias N. Bonten; Martijn H. Rutten; Patricia M. Stassen; Bas L. J. H. Kietselaer; Geert-Jan Dinant; Jochen W.L. Cals

doi:10.1370/afm.2401

Article Figures & Data

Figures

Tables
Additional Files

Download figure
Open in new tab
Download powerpoint
Figure 1
Implementation of the flash-mob method in the present study.
^a Specialized FPs receive a special 2-year training in heart and vascular disease or acute medicine.
Download figure
Open in new tab
Download powerpoint
Figure 2
Patient inclusion and follow-up.
Download figure
Open in new tab
Download powerpoint
Figure 3
ROC curves for FP assessment of ACS probability and the Marburg Heart Score in patients referred for suspected ACS (n = 186).
ACS = acute coronary syndrome; AUC = area under the curve; FP = family physician; ROC = receiver operating characteristic.

View popup

Table 1

Components of the Marburg Heart Score^1–3

Score Component	Assigned Points
Age/sex (female aged ≥65 y, male aged ≥55 y)	1
Known clinical vascular disease^a	1
Patient assumes cardiac origin of pain	1
Pain worse with exercise	1
Pain not reproducible by palpation	1

↵a Including coronary heart disease, cerebrovascular disease, or peripheral artery disease.
Note: 1 point is assigned to each score variable; 3 different risk categories are derived (low risk = 0-2 points; intermediate risk = 3 points; high risk = 4-5 points).

View popup

Table 2

Patient Characteristics, Investigated Predictors, and Univariate Analyses of Possible ACS Predictors

	Total	ACS	No ACS	Odds Ratio (95% CI)	P Value
Mean age, y (SD)	64 (13.3)	67 (12.6)	63 (13.3)
Male, n/N (%)	127/243 (52.3)	34/45 (75.6)	93/198 (47.0)	3.49 (1.67-7.28)	<.01
Sex-adjusted age; female aged ≥65 y, male aged ≥55 y, n/N (%)^a	153/243 (63.0)	36/45 (80.0)	117/198 (59.1)	2.77 (1.27-6.06)	.01
Duration of symptoms, n/N (%)
<1 h	23/243 (9.5)	3/45 (6.7)	20/198 (10.1)
1-24 h	138/243 (56.8)	23/45 (51.1)	115/198 (58.1)
>24 h	82/243 (33.7)	19/45 (42.2)	63/198 (31.8)
FP immediately suspected a serious condition, n/N (%)	120/243 (49.4)	26/45 (57.8)	94/198 (47.5)	1.51 (0.79-2.91)	.21
History of clinical vascular disease, n/N (%)a	106/243 (43.6)	22/45 (48.9)	84/198 (42.4)	1.30 (0.68-2.48)	.43
Patient assumes cardiac origin of pain, n/N (%)^a	136/242 (56.2)	28/45 (62.2)	108/197 (54.8)	1.36 (0.70-2.64)	.37
Chest pain, n/N (%)	215/243 (88.5)	39/45 (86.7)	176/198 (88.9)	0.81 (0.31-2.14)	.67
Pain worse with exercise, n/N (%)^a	86/188 (45.7)	17/36 (47.2)	69/152 (45.4)	1.08 (0.52-2.23)	.84
Pain feels like pressure, n/N (%)	185/212 (87.3)	31/39 (79.5)	154/173 (89.0)	0.48 (0.19-1.19)	.11
Pain not reproducible by palpation, n/N (%)^a	179/210 (85.2)	36/39 (92.3)	143/171 (83.6)	2.35 (0.68-8.16)	.17
ECG performed, n/N (%)	115/239 (48.1)	23/44 (52.3)	92/195 (47.2)
ECG shows ischemic changes, n/N (%)	37/115 (32.2)	17/23 (73.9)	20/92 (21.7)	10.20 (3.55-29.28)	<.01
FP assessment of ACS probability, median (IQR)	6 (3)	7 (3)	6 (3)
Marburg Heart Score,^b median (IQR)	3 (2)	4 (2)	3 (2)

ACS = acute coronary syndrome; ECG = electrocardiography; FP = family physician; IQR = interquartile range.
↵a Component of the Marburg Heart Score.
↵b See Table 1 for all of the components of the Marburg Heart Score.

View popup

Table 3

Test Characteristics of the Marburg Heart Score and the FP Assessment

Variable	ACS	No ACS	Total
MHS using a cut-off value of ≤2
MHS >2	27	84	111
MHS ≤2	9	66	75
Total	36	150	186

Sensitivity: 75.0% (95% CI, 57.5-87.3)	Incidence ACS: 19.4% (95% CI, 14.1-25.9)
Specificity: 44.0% (95% CI, 36.0-52.3)	Incidence No ACS: 80.6% (95% CI, 74.1-85.9)
PPV: 24.3% (95% CI, 16.9-33.6)	OR: 2.36 (95% CI, 1.04-5.35)
NPV: 88.0% (95% CI, 78.0-94.0)	χ²: 0.04

MHS using a cut-off value of ≤1
MHS >1	34	126	160
MHS ≤1	2	24	26
Total	36	150	186

Sensitivity: 94.4% (95% CI, 80.0-99.0)	Incidence ACS: 19.4% (95% CI, 14.1-25.9)
Specificity: 16.0% (95% CI, 10.7-23.1)	Incidence No ACS: 80.6% (95% CI, 74.1-85.9)
PPV: 21.3% (95% CI, 15.4-28.6)	OR: 3.24 (95% CI, 0.73-14.39)
NPV: 92.3% (95% CI, 73.4-98.7)	χ²: 0.11

FP probability assessment using a cut-off value of ≤5
FP probability assessment high (>5)	39	116	155
FP probability assessment low (≤5)	6	82	88
Total	45	198	243

Sensitivity: 86.7% (95% CI, 72.5-94.5)	Incidence ACS: 18.5% (95% CI, 14.0-24.1)
Specificity: 41.4% (95% CI, 34.5-48.6)	Incidence No ACS: 81.5% (95% CI, 75.9-86.0)
PPV: 25.2% (95% CI, 18.7-32.9)	OR: 4.60 (95% CI, 1.86-11.36)
NPV: 93.2% (95% CI, 85.2-97.2)	χ²: <0.01

Combined approach using the MHS and the FP assessments of probability
MHS + FP probability assessment, 1-2^a	36	115	151
MHS + FP probability assessment, 0^b	0	35	35
Total	36	150	186

Sensitivity: 100% (95% CI, 88.0-100)	Incidence ACS: 19.4% (95% CI, 14.1-25.9)
Specificity: 23.3% (95% CI, 17.0-31.1)	Incidence No ACS: 80.6% (95% CI, 74.1-85.9)
PPV: 23.8% (95% CI, 17.5-31.6)
NPV: 100% (95% CI, 87.7-100)

ACS = acute coronary syndrome; FP = family physician; MHS = Marburg Heart Score; NPV = negative predictive value; OR = odds ratio; PPV = positive predictive value.
↵a Either the MHS, the FP assessment, or both were positive using an MHS cut-off value of ≤2 and an FP assessment cut-off value of ≤5.
↵b Both the MHS and the FP assessments were negative.

Additional Files

Figures
Tables

The Article in Brief

A Nationwide Flash-Mob Study for Suspected Acute Coronary Syndrome

Jochen W.L. Cals , and colleagues

Background The dual aim of this study was to evaluate the Marburg Heart Score, a clinical decision rule, and/or develop an adapted clinical decision rule for family practitioners to safely rule out acute coronary syndrome in patients referred to secondary care for suspected ACS; and also to evaluate the feasibility of using a "flash mob" method, an innovative new study design, for large scale research in family medicine.

What This Study Found This study finds that, in emergency care, acute coronary syndrome cannot be safely ruled out using the Marburg Heart Score or the family physicians' clinical assessment. In a period of only 2 weeks, researchers at Maastricht University collected data on 258 ACS-suspected patients by mobilizing 1 in 5 family physicians throughout the Netherlands to participate in the study. This mobilization was done by enlisting ambassadors among the FP community in the Netherlands who then spread the word through traditional professional and social networks. The study found that among 243 patients receiving a final diagnosis, 45 (18.5%) were diagnosed with acute coronary syndrome. Sensitivity for the FP rating was 86.7% and sensitivity for the MHS was 94.4%.

Implications

While large, prospective studies can be time consuming and costly, this innovative flash mob method of research, named after the large-scale public collaborations/gatherings driven by social media, allowed for the fast investigation of one simple question on a large scale in a short timeframe.