Table 2.

US 1989, 1992, 1995 and 2006 cocaine essential chemical regulation impacts on cocaine seizure amount^a, price^b and purity^c in the conterminous United States (US): model parameter estimates.

		Purity‐adjusted seizure amount			Purity‐adjusted price			Purity
	Parameter	Est	(95% CI)	% change	Est.	(95% CI)	% change	Est.	(95% CI)	% change
Series period (January 1987–December 2000)
1989 potassium permanganate/solvents regulation	ω 1	−0.32**	(−0.59, −0.06)	−28%	15.52**	(2.73, 28.31)	+36%	−3.21**	(−6.08, −0.33)	−4%
	δ 1	–d	–		0.57**	(0.13, 1.01)		–	–
1992 acids regulation	ω 2	−0.34**	(−0.60, −0.07)	−29%	5.56	(−8.62, 19.74)		0.16	(−2.72, 3.04)
1995 MIBK regulation	ω 3	−0.08	(−0.35, 0.18)		21.13***	(7.02, 35.24)	+25%	−1.68	(−4.57, 1.20)
ARIMA	θ 1	0.65***	(0.54, 0.77)		0.29***	(0.14, 0.44)		0.14*	(−0.01, 0.30)
Series period (January 2000–April 2011)
2006 sodium permanganate regulation	ω 1	−0.25**	(−0.47, −0.03)	−22%	6.18**	(0.97, 11.39)	+100%	−1.80**	(−3.58, −0.02)	−35%
	δ 1	–	–		0.91***	(0.81, 1.00)		0.93***	(0.85, 1.01)
ARIMA	θ 1	0.70***	(0.57, 0.82)		0.47***	(0.31, 0.62)		0.18**	(0.01, 0.34)
	ϕ 12	0.30***	(0.14, 0.46)		–	–		–	–

The seizure amount series was log‐transformed. Est. = estimate; CI = confidence interval; ω _i = intervention impact; δ _i = impact rate; θ ₁ = non‐seasonal moving average parameter; ϕ ₁₂ = seasonal autoregressive parameter; MIBK = methyl isobutyl ketone; ARIMA = autoregressive integrated moving average. For the 1992 and 1995 regulations, gradual impacts were not found; consequently, δ ₂ and δ ₃ are not listed.

^aMonthly totals of pure cocaine in seizures < 6000 g gross weight;

^bmedian price per pure gram adjusted to 2010 US dollars: all seizures;

^cmedian purity: all seizures;

^ddashes (–) indicate that the parameter estimate was not in the model.

^* P < 0.10;

^** P < 0.05;

^*** P < 0.01.