Chest pain: tests made patients feel better.

Clinical bottom line (level 1b)

  1. Patients with chest pain who were at low-risk for ischaemic heart disease, who received basic investigations had improved activity at 3 weeks (NNT = 4 at 3 weeks) . Fewer felt unwell (NNT = 7 at 3 weeks) , and more considered their care was better than usual (NNT = 4 at 3 weeks) .
Sox et al: Annals of Internal Medicine 1981; 95: 680-685
Expires March 2003

The study

Single-blinded concealed randomised trial without intention-to-treat
Setting: VA centre, USA

186 patients (aged ~50% <50 years old, 98% male) with chest pain considered low-risk for ischaemic heart disease (based on a clinical prediction rule)
Note:
  • Patients answered a questionnaire on how they felt at 3 weeks and 5 months.


Control Group: (n = 93, 84 analysed): 12-lead ECG and CK
Experimental Group: (n = 93, 87 analysed): no tests performed

91% followed for 5 months

The evidence

Outcome Time to outcome CEREERRRR
(95% CI)		ARR
(95% CI)		NNT
(95% CI)
activity same or less than before 3 weeks 17
(20.2%)		 39
(44.8%)		 -122%
(-260% to -36%)		 -24.6%
(-38.1% to -11.1%)		 -4
(-9 to -3)
'feel sick all the time' 3 weeks 9
(10.7%)		 23
(26.4%)		 -147%
(-402% to -21%)		 -15.7%
(-27.1% to -4.34%)		 -6
(-23 to -4)
pain gone 3 weeks 34
(40.5%)		 33
(37.9%)		 6%
(-36% to 35%)		 2.55%
(-12.1% to 17.2%)		 39
(NNT = 6 to infinity;
NNH = 8 to infinity)
care 'better than usual' 3 weeks 48
(57.1%)		 27
(31.0%)		 46%
(22% to 62%)		 26.1%
(11.7% to 40.5%)		 4
(2 to 9)
activity same or less than before 5 months 26
(31.0%)		 23
(26.4%)		 15%
(-37% to 47%)		 4.52%
(-9.03% to 18.1%)		 22
(NNT = 6 to infinity;
NNH = 11 to infinity)
pain gone 5 months 55
(65.5%)		 49
(56.3%)		 14%
(-10% to 32%)		 9.15%
(-5.41% to 23.7%)		 11
(NNT = 4 to infinity;
NNH = 18 to infinity)

Comments

  1. How much do these results reflect the group studied? Mainly men and mainly ex-US soldiers.
  2. Patients in both groups were still equally worried about heart disease whether tested or not.

Citation

  1. Sox HC, Margulies I, Sox CH: Psychologically mediated effects of diagnostic tests. Annals of Internal Medicine 1981; 95: 680-685
Search Terms: chest pain in Cochrane
Contributor: Chris Ball and Clare Wotton, March 2000
Reviewer:

Clinical Question.
    Patient chest pain
    Intervention or Exposure tests performed
    Comparison no tests performed
    Outcome how patients feel