Quantifying the defensive medicine contribution to primary care costs

Figures & data

Table 1. Physicians characteristics.

Physician characteristic	Total n (%)	Highly defensive physicians, n (%)	Less defensive physicians, n (%)	p-Value
Sex				0.67
Women	10 (48)	6 (55)	4 (40)
Men	11 (52)	5 (45)	6 (60)
Years in practice				1.00
<5	4 (19)	2 (18)	2 (20)
5–10	4 (19)	2 (18)	2 (20)
11–20	8 (38)	4 (36)	4 (40)
>20	5 (24)	3 (27)	2 (20)
Medical school				0.36
US	15 (71)	9 (82)	6 (60)
Non-US	6 (29)	2 (18)	4 (40)
Residency				0.01
Academic	14 (70)	4 (40)	10 (100)
Community	6 (30)	6 (60)	0 (0)
How worried are you about being sued?				0.47
Not at all	2 (10)	0 (0)	2 (20)
A little/some	16 (80)	8 (80)	8 (80)
Somewhat	1 (5)	1 (10)	0 (0)
A lot	1 (5)	1 (10)	0 (0)
Ever been sued				0.30
No	15 (75)	6 (60)	9 (90)
Yes	5 (25)	4 (40)	1 (10)
Know someone who has been sued				0.47
No	2 (10)	0 (0)	2 (20)
Yes	18 (90)	10 (100)	8 (80)
How much do you feel defensive medicine actually decreases your risks of ever being sued?				0.17
Not at all	3 (15)	0 (0)	3 (30)
A little	10 (50)	5 (50)	5 (50)
Somewhat	7 (35)	5 (50)	2 (20)
A lot	0 (0)	0 (0)	0 (0)
What do you believe is the probability you will be sued in the next 10 years?				0.21
<1%	3 (15)	0 (0)	3 (30)
1–9%	15 (75)	9 (90)	6 (60)
10–19%	2 (10)	1 (10)	1 (10)
≥20%	0 (0)	0 (0)	0 (0)

Table 2. Summary of outcomes.

Outcomes	All physicians (n = 21)	Highly defensive physicians* (n = 11)	Less defensive physicians* (n = 10)	p-Value
Total number of orders†	58.8 (23.4)	53.5 (23.0)	64.5 (23.6)	0.30
Total cost,‡ $	3983.6 (1741.5)	3814.1 (2016.9)	4170.1 (1465.1)	0.65
Orders† per patient‡ per visit	3.6 (1.0)	3.4 (0.9)	3.7 (1.1)	0.41
Cost† per order per patient,‡ $	72.6 (18.5)	74.1 (19.2)	70.9 (18.7)	0.70
Total cost^† per patient‡ per visit, $	241.8 (77.6)	239.9 (93.2)	243.9 (61.0)	0.91

*Highly defensive and less defensive physicians are those with defensiveness scores above and below the median of 0.018, respectively.

†Mean (standard deviation).

‡For each outcome defined per patient, we averaged it within patients first and then summarized the averaged data by providers.

Table 3. Association between physician characteristics and defensive orders. The logistic model contains a random effect for the provider to account for the possible correlation among repeated orders placed by each physician.

Physician characteristic	Odds ratio (95% CI)	p-Value
Gender
Female	1.00
Male	0.73 (0.23–2.34)	0.60
Years in practice
<5	1.00
5–10	3.41 (0.64–18.07)	0.15
11–20	3.38 (0.77–14.75)	0.11
>20	1.06 (0.20–5.76)	0.94
Medical school
US	1.00
Non-US	0.37 (0.11–1.24)	0.11
Residency
Academic	1.00
Community	4.29 (1.55–11.86)	0.01
How worried are you about being sued?
Not at all	1.00
A little/some	4.26 (0.47–38.80)	0.20
Somewhat	2.97 (0.11–77.04)	0.51
A lot	5.78 (0.26–129.81)	0.27
Ever been sued
No	1.00
Yes	1.61 (0.45–5.85)	0.47
Know someone who has been sued
No	1.00
Yes	1.77 (0.26–12.13)	0.56
How much do you feel defensive medicine actually decreases your risks of ever being sued?
Not at all	1.00
A little	2.77 (0.44–17.57)	0.28
Somewhat	3.99 (0.60–26.59)	0.15
A lot	–	–
What do you believe is the probability you will be sued in the next 10 years?
<1%	1.00
1–9%	4.45 (0.77–25.78)	0.10
10–19%	2.56 (0.25–26.47)	0.43
≥20%	–	–

Figure 1. Proportion of orders deemed to be defensive by order type.

Table 4. Association between patient characteristics and defensive orders. The logistic model contains a random effect for provider to account for the possible correlation among repeated orders placed by each physician.

Patient characteristics	Proportion of defensive orders (%)	Odds ratio (95% CI)	p-Value
Gender
Male	8.5	1.00
Female	11.3	1.69 (1.06–2.70)	0.03
Age
Per 1 year		1.00 (0.99–1.01)	0.92
Insurance
Private	9.7	1.00
Medicare	11.5	0.95 (0.61–1.49)	0.83
Medicaid	20.6	4.98 (1.17–21.26)	0.03
Self-pay	4.7	0.55 (0.21–1.42)	0.22
Order category
Vaccine	1.6	1.00
Consult	11.8	8.25 (1.02–66.92)	0.048
Imaging	18.4	14.19 (1.76–114.27)	0.013
Lab	8.0	5.27 (0.68–41.13)	0.11
Equipment	0	0	0.99
Cardiac	43.5	70.02 (7.33–669.17)	<0.001
Procedure	10.4	8.92 (0.93–86.09)	0.058