Skip to main content

Advertisement

Table 1 Comparison of Covariate Balance between NSAIDs and Acetaminophen before and after propensity score matching

From: Comparative effects of non-steroidal anti-inflammatory drugs (NSAIDs) on blood pressure in patients with hypertension

Variable Sample NSAIDs* Acetaminophen* P Value** Standardized Difference Bias Reduction (%)
Age (yrs) mean Unmatched 55 60 <.001 −42.0  
  Matched 56 57 0.119 −6.0 86%
Gender:       
Female Unmatched 70 70 0.687 1.3  
  Matched 72 70 0.157 5.5 −323%
Race:       
African American Unmatched 58 63 0.003 −9.4  
  Matched 63 61 0.353 3.6 62%
Other Unmatched 5 3 0.007 7.5  
  Matched 4 3 0.837 0.8 89%
Baseline systolic blood pressure (mmHg) mean Unmatched 139 141 0.006 −8.8  
  Matched 140 140 0.949 −0.2 97%
Time from baseline SBP to index:       
≤ 7 days Unmatched 53 65 <.001 −23.8  
  Matched 62 61 0.596 2.0 91%
> 7 days and ≤ 30 days Unmatched 14 10 <.001 11.6  
  Matched 10 11 0.380 −3.4 71%
> 30 days Unmatched 33 25 <.001 17.4  
  Matched 28 28 0.975 0.1 99%
Year of index date:       
1993 - 1996 Unmatched 50 58 <.001 −17.1  
  Matched 57 56 0.713 1.4 92%
1997-2002 Unmatched 39 40 0.477 −2.3  
  Matched 41 42 0.741 −1.3 44%
2003 - 2006 Unmatched 11 1 <.001 40.8  
  Matched 2 2 0.881 −0.6 99%
Diagnosis of:       
Osteoarthritis Unmatched 22 23 0.768 −0.9  
  Matched 21 22 0.295 −4.0 −327%
Rheumatoid Arthritis Unmatched 3 3 0.381 2.8  
  Matched 3 3 0.819 0.9 69%
Renal Insufficiency Unmatched 3 8 <.001 −24.5  
  Matched 4 4 0.922 0.4 98%
Cirrhosis with Ascites Unmatched 0.4 1 0.343 −3.0  
  Matched 1 1 0.807 −0.9 69%
Systemic Lupus Erythematosus Unmatched 1 1 0.277 3.5  
  Matched 1 1 0.465 −2.8 20%
Diabetes Unmatched 28 35 <.001 −14  
  Matched 28 31 0.124 −5.9 58%
Congestive Heart Failure Unmatched 11 19 <.001 −20.9  
  Matched 14 14 0.651 −1.7 92%
Coronary Artery Disease or History of Myocardial Infarction Unmatched 13 19 <.001 −16.0  
  Matched 14 15 0.266 −4.3 73%
Stroke Unmatched 8 12 <.001 −15.4  
  Matched 9 9 0.784 −1.1 93%
Arrhythmia Unmatched 1 2 0.055 −6.1  
  Matched 1 1 0.489 −2.7 56%
Asthma or Chronic Obstructive Pulmonary Disease Unmatched 17 19 0.121 −5.0  
  Matched 18 19 0.420 −3.1 37%
Medications:       
ACE-I or Angiotensin II blocker Unmatched 36 43 <.001 −15.7  
  Matched 37 38 0.594 −2.1 87%
Beta- Blocker Unmatched 18 17 0.270 3.5  
  Matched 15 17 0.338 −3.7 −4%
Calcium Channel Blocker Unmatched 34 38 0.003 −9.7  
  Matched 36 36 0.829 −0.8 91%
Diuretic Unmatched 44 48 0.005 −9.0  
  Matched 46 45 0.921 0.4 96%
Other BP medications Unmatched 7 12 <.001 −15.2  
  Matched 9 9 0.835 −0.8 95%
Oral high dose glucocorticoid*** Unmatched 1 1 0.068 −5.8  
  Matched 1 1 0.850 −0.7 87%
Oral Contraceptives Unmatched 1 1 0.528 −2.0  
  Matched 1 1 0.998 0.0 100%
Venlafaxine Unmatched 1 0.3 0.075 5.8  
  Matched 1 0.4 0.782 1.1 82%
Adherence to antihypertensive medications:       
MPR ≥ 80% Unmatched 69 74 <.001 −10.6  
  Matched 72 70 0.320 3.8 64%
MPR < 80% Unmatched 13 12 0.262 3.6  
  Matched 12 13 0.159 −5.4 −51%
Not using antihypertensives (reference group) Unmatched 18 14 0.002 9.8  
  Matched 16 16 0.965 0.2 98%
Exposure to index drug:       
MPR > 80% Unmatched 32 24 <.001 19.5  
  Matched 26 27 0.750 −1.2 94%
MPR 20–80 % Unmatched 44 43 0.278 3.5  
  Matched 44 43 0.859 0.7 80%
MPR < 20% Unmatched 23 34 <.001 −23.3  
  Matched 30 30 0.909 0.4 98%
Number of refills per month:       
≥ 1 refills Unmatched 52 36 <.001 30.8  
  Matched 44 43 0.301 4.0 87%
  1. * % unless indicated as mean. Because of rounding values may not add to 100 %.
  2. ** P-value of t-tests for continuous variables and chi-square tests for categorical variables.
  3. *** High dose was defined as ≥10mg for prednisonse, ≥50mg for cortisone, and ≥1.5 mg for dexamethasone.
  4. Standardized difference: 100 ( χ ¯ treated χ ¯ control)/√ {(s2 treated + s2 control)/2}. A positive value means the treated group is higher in % (or mean) compared to the control group and negative value means the control is higher than the treated.
  5. Bias reduction (%) =1- {|Standardized difference matched|/| Standardized difference unmatched |} x 100. A positive value means bias is reduced by propensity score matching and negative means bias increased.
  6. ACE-I: Angiotensin converting enzyme inhibitor; MPR: Medication Possession Ratio; SBP: Systolic blood pressure. Unmatched: all patients before propensity score matching, N=3,928 (2,181 NSAIDs and 1,747 acetaminophen). Matched: only matched patients, N=2,680 (1,340 in each group).