Prevalence and predictors of over-the-counter medication use among pregnant women: a cross-sectional study in the Netherlands

Table 5 Scores, predictive probabilities and the percentage of women in each category

Score	Mean predictive probability	Sensitivity	Specificity	Number of persons	Number of persons (%)	Cumulative percent number of persons
13	0.574			1	0.1	100.0
12	-			-	-	-
11	0.456			1	0.1	99.9
10	0.411	0.027	0.989	13	1.1	99.8
9	0.349	0.082	0.974	23	2.0	98.7
8	0.301	0.122	0.956	24	2.1	96.7
7	0.258	0.170	0.939	25	2.2	94.6
6	0.217	0.286	0.897	59	5.1	92.5
5	0.182	0.374	0.841	70	6.1	87.4
4	0.155	0.497	0.740	120	10.4	81.3
3	0.124	0.599	0.634	122	10.5	71.0
2	0.102	0.857	0.342	333	28.8	60.4
1	0.081	0.912	0.256	94	8.1	31.6
0	0.066			272	23.5	23.5

The score is the outcome of the risk score formula:
Risk score = 2 x prescription drug + 2 x nulliparity + 4 x comorbidity + 2 x BMI + number of visits GP.
Prescription drug: fill out 1 if any kind of prescription drug is used.
Nulliparity: fill out 1 if the woman is nulliparous.
Comorbidity: fill out 1 if any kind of comorbidity is reported.
BMI: fill out 1 if BMI is between 26 and 30 kg/m².
Number of visits GP: fill out the number of visits to the GP in the last 10 months. If the woman visited the GP more than three times, fill out 3.
The mean predictive probability is the chance that a women with that score uses OTC-medication. The number of persons are the number of persons in our database having a particular score.

ISSN: 1471-2458