Investigating over-the-counter oral analgesics

Clinical bottom line
There is good evidence supporting the efficacy of standard doses of aspirin, paracetamol, ibuprofen, naproxen, and diclofenac, all of which are available as over-the-counter (OTC) medicines in some part of the world. There is no good evidence for most branded combination products, though it is likely that additional analgesic effect is produced by codeine.

Background

A wide variety of over-the-counter (OTC) analgesics are available to buy, but the amount of high quality information about these treatments is limited. We set out to find evidence for the efficacy of a range of OTC analgesics, available in various parts of the world, in standard acute pain trials. Specifically we were looking for single dose data from 4-6 hour trials in post-operative pain models, and reporting standard outcomes.

Clinical trials measuring the efficacy of analgesics in acute pain have been standardised over many years. Trials have to be randomised and double blind. Typically, in the first few hours or days after an operation, patients develop pain that is moderate to severe in intensity, and will then be given the test analgesic or placebo. Pain is measured using standard pain intensity scales immediately before the intervention, and then using pain intensity and pain relief scales over the following 4 to 6 hours for shorter acting drugs. Pain relief of half the maximum possible pain relief or better (at least 50% pain relief) is typically regarded as a clinically useful outcome. For patients given rescue medication it is usual for no additional pain measurements to be made, and for all subsequent measures to be recorded as initial pain intensity or baseline (zero) pain relief (baseline observation carried forward). This process ensures that analgesia from the rescue medication is not wrongly ascribed to the test intervention. In some trials the last observation is carried forward, which gives an inflated response for the test intervention compared to placebo, but the effect has been shown to be negligible over four to six hours.

Single dose trials in acute pain are commonly short in duration, rarely lasting longer than 12 hours, allowing no reliable conclusions to be drawn about safety. To show that the analgesic is working it is necessary to use placebo. There are clear ethical considerations in doing this. These ethical considerations are answered by using acute pain situations where the pain is expected to go away, and by providing additional analgesia, commonly called rescue analgesia, if the pain has not diminished after about an hour. This is reasonable, because not all participants given an analgesic will have significant pain relief. Approximately 18% of participants given placebo will have significant pain relief, and up to 50% may have inadequate analgesia with active medicines.

Systematic review and methods

For references to methods used, refer to Moore et al., Bandolier's Little Book of Pain, Oxford University Press, 2003.

We searched PubMed, Cochrane Central Library, and our own in-house databases in pain research for any double-blind, randomised controlled trials reporting pain relief, pain intensity, or patient global evaluation of efficacy as outcomes over 4-6 hours for single dose analgesic versus placebo. The search terms used included both trade names and generic names of the individual analgesic constituents, including combinations where appropriate. It is not likely that all OTC analgesics have been included, since sources for OTC analgesic names and availability are not easy to come by, and may change from time to time. OTC analgesic combinations, in particular, may change. The approach, therefore, was to work with combinations of drugs and doses of the combinations that appeared to be current in 2009.

From these trials we extracted outcome data, including pain relief measured as a TOTPAR (total pain relief) at 4 or 6 hours, and pain intensity measured as a SPID (summed pain intensity difference) at 4 or 6 hours. Mean TOTPAR or SPID values, for both the active analgesic and placebo, were then converted to %maxTOTPAR or %maxSPID by division into the calculated maximum value. The proportion of participants in each treatment group who achieved at least 50%maxTOTPAR was calculated using verified equations, and these proportions converted into the number of participants achieving at least 50%maxTOTPAR by multiplying by the total number of participants in the treatment group. Information on the number of participants with at least 50%maxTOTPAR for active treatment and placebo was then used to calculate relative benefit (RB) and number-needed-to-treat-to-benefit (NNT).

Results

One hundred and twenty five RCTs were retrieved that matched the search criteria. After closer scrutiny, six head-to-head comparative trials were excluded due to lack of a placebo control, and two trials were excluded due to lack of analysable data. The remaining one hundred and seventeen trials were randomised, double blind and placebo controlled and were included in the efficacy analysis. The studies involved a mixture of dental pain and episiotomy pain.

The overall standard and quantity of data available was poor, particularly for studies specifically using the trade name over-the-counter analgesics. To compensate for this we have included data on the equivalent dose generic named analgesics and their combinations. For some of the test analgesics (Anadin Extra, Askit, Codis, Dispirin, Dispirin Extra, Panadeine 15, Paracodol, Paramol, Pentalgin H, Sedalgin-neo, Solpadeine Max) no useable data could be found. In many cases, particularly those combination analgesics including codeine, this was due to differences in the doses of the constituent analgesics used in the available trials as compared with the over-the-counter versions. In general, over-the-counter analgesics containing codeine tended to use significantly lower doses of codeine and higher doses of other constituents; presumably to minimise codeine-related side effects. Information on combinations of paracetamol and ibuprofen is included since these newer combinations are likely to appear as OTC analgesics in several parts of the world. Table 1 gives information about the included studies.

Table 1: Details of available data

Drug	Details of available data	References of included studies
Anadin Extra	We found no trials comparing Anadin Extra (or a generic combination analgesic containing paracetamol, aspirin and caffeine in similar doses) to placebo	N/A
Askit	We found no trials comparing Askit (or a generic combination analgesic containing aspirin, caffeine and aloxiprin in similar doses) to placebo	N/A
Aspirine	We found two trials: Forbes et al. 1990 and Rubin et al. 1984 comparing a generic combination of aspirin and caffeine (ASA 650mg/caffeine 65mg in Forbes et al. and ASA 800mg/caffeine 65mg in Rubin et al.) against placebo. Both trials were relatively small and used different pain types: Forbes et al. (n=141) in dental and Rubin et al. (n=230) in episiotomy. The results reflect this with Forbes et al. reporting the % of patients achieving 50% pain relief on the active treatment as 27% and on the placebo as 1%; while Rubin et al. report 86% on the active treatment and 48% on the placebo	Forbes JA. Pharmacotherapy. 1990; 10(6):387-93 Rubin A. J Int Med Res. 1984; 12(6):338-45
Aspro Clear	We found seven trials in a Cochrane review of single dose oral aspirin for acute pain (Edwards et al. 2000 - currently undergoing in-house update) comparing aspirin in any formulation (ASA 1000mg) against placebo	Edwards JE. Cochrane Database Syst Rev. 2000;(2):CD002067
Codis	We found no trials comparing Codis (or a generic combination analgesic containing aspirin and codeine in similar doses) to placebo	N/A
Cuprofen Plus	We found two trials: Cater et al. 1985 and Norman et al. 1985 comparing a generic combination of ibuprofen and caffeine (IBU 400mg/COD 30mg) against placebo. Both trials reported pain following episiotomy with similar results	Cater M. Clin Ther. 1985; 7(4):442-7 Norman SL. Clin Ther. 1985; 7(5):549-54
Disprin	We found no trials comparing Dispirin (or a generic formulation of aspirin in a similar dose) to placebo	N/A
Disprin Extra	We found no trials comparing Dispirin Extra (or a generic combination of aspirin and paracetamol in similar doses) to placebo	N/A
Feminax Ultra	We found five trials in an up-to-date Cochrane review of single dose oral naproxen for acute pain (Derry et al. 2009) comparing naproxen or naproxen sodium (NAPROX 500mg or NAPROX SODIUM 550mg) against placebo	Derry C. Cochrane Database Syst Rev. 2009 Jan 21;(1);CD004234
Mersyndol	We found one trial: Margarone et al. 1995 comparing Mersyndol against placebo. The trial reported pain following dental surgery	Margarone JE. Clin Pharmacol Ther. 1995 Oct; 58(4):453-8
Nurofen	We found 61 trials in an up-to-date Cochrane review of single dose oral ibuprofen for acute pain (Derry et al. 2009) comparing a generic formulation of ibuprofen (IBU 400mg) against placebo	Derry C. Cochrane Database Syst Rev. 2009 Jul 8;(3):CD001548
Panadeine 15	We found no trials comparing Panadeine 15 (or a generic combination of paracetamol and codeine in similar doses) to placebo	N/A
Panadol	We found 28 trials in an up-to-date Cochrane review of single dose oral paracetamol for acute pain (Toms et al. 2008) comparing a generic formulation of paracetamol (PARA 1000mg) against placebo	Toms L. Cochrane Database Syst Rev. 2008 Oct 8;(4):CD004602
Panadol Extra	We found one trial: Winter et al. 1983 comparing a generic combination of paracetamol and caffeine (PARA 1000mg/CAF 130mg) against placebo. The trial reported pain following dental surgery	Winter L Jr. Current Therapeutic Research. 1983 Jan; 33(1):115-122
Paracodol	We found no trials comparing Paracodol (or a generic combination of paracetamol and codeine in similar doses) to placebo	N/A
Paramol	We found no trials comparing Paramol (or a generic combination of paracetamol and dihydrocodeine tartrate in similar doses) to placebo	N/A
Pentalgin H	We found no trials comparing Pentalgin H (or a generic combination of naproxen, codeine, caffeine, dipyrone and phenobarbitol in similar doses) to placebo	N/A
Saridon	We found one trial: Kiersch et al. 2002 comparing Saridon against placebo. The trial reported pain following dental surgery	Kiersch TA. Curr Med Res Opin. 2002; 18(1):18-25
Sedalgin-neo	We found no trials comparing Sedalgin-neo (or a generic combination of paracetamol, caffeine, codeine, dipyrone and phenobarbitol in similar doses) to placebo	N/A
Solpadeine Max	We found no trials comparing Solpadeine Max (or a generic combination of paracetamol and codeine in similar doses) to placebo	N/A
Solpadeine Plus	We found one trial: Cooper et al. 1986 comparing a generic combination of paracetamol, codeine and caffeine (PARA 1000mg/COD 16mg/CAF 30mg) against placebo. The trial reported pain following dental surgery	Cooper SA. Anesth Prog. 1986 May-Jun; 33(3):139-42
Voltarol	We found four trials in an up-to-date Cochrane review of single dose oral diclofenac for acute pain (Derry et al. 2009) comparing all generic formulations of diclofenac (DICLO 25mg) against placebo	Derry P. Cochrane Database Syst Rev. 2009 Apr 15;(2):CD004768

Table 2 summarises data available for each of the analgesics along with its calculated relative benefit (RB) and number-needed-to-treat-to-benefit (NNT). Para = paracetamol, Asa - aspirin, Caf = caffeine, Cod = codeine, Naprox = naproxen, Diclo = diclofenac, Ibu = ibuprofen

Drug	Constituents	Number of Trials	Number of Patients	Percent with Active	Percent with Control	RB (95% CI)	NNT (95% CI)
Anadin Extra	Para400 + Asa600 + Caf90	0
Askit	Asa530 + Caf110 + Aloxiprin140	0
Aspirine	Asa650 + Caf65	2	371	65	28	2.3 (1.8 - 3.0)	2.7 (2.2 - 3.7)
		Forbes 1990	141	17	0	39.8 (2.4 - 648)	3.9 (2.7 - 6.7)
		Rubin 1984	230	86	48	1.8 (1.5 - 2.2)	2.6 (2.0 - 3.7)
Aspro Clear	Asa1000	7	679	43	16	2.6 (2.0 - 3.5)	3.7 (3.0 - 5.0)
Codis	Asa1000 + Cod base 16	0
Cuprofen Plus	Ibu400 + Cod base 20	2	167	55	31	1.8 (1.2 - 2.6)	4.1 (2.6 - 10.3)
Disprin	Asa900	0
Disprin Extra	Asa600 + Para400	0
		Margarone 1995	76	21	8	2.7 (0.8 - 9.3)	7.6
Feminax Ultra	Naprox500	9	784	52	15	3.4 (2.7 - 4.4)	2.7 (2.3 - 3.2)
		Winter 1983	81	48	22	2.2 (1.1 - 4.2)	3.9 (2.2 - 18.0)
Mersyndol	Para1000 + Cod base15 + Doxylamine succinate10	1	76	21	8	2.7 (0.8 - 9.3)	Not Calculated
Nurofen	Ibu400	61	6475	54	14	4.0 (3.6 - 4.4)	2.5 (2.4 - 2.6)
Panadeine 15	Para1000 + Cod base23	0
Panadol	Para1000	28	3232	46	18	2.5 (2.2 - 2.9)	3.6 (3.2 - 4.0)
		Cooper 1986	61	29	4	6.2 (0.9 - 45.0)	4.2 (2.5 - 14.2)
Panadol Extra	Para1000 + Caf130	1	81	48	22	2.2 (1.1 - 4.2)	3.9 (2.2 - 18.0)
Paracodol	Para1000 + Cod base13	0
Paramol	Para1000 + Dihydrocodeine tartarate15	0
Pentalgin H	Naprox100 + Cod base8 + Caf50 + Dipyrone300 + Phenobarbitol15	0
Saridon	Para500 + Caf100 + Propifenazone300	1	301	23	2	9.2 (1.3 - 64.5)	4.9 (3.6 - 7.4)
Sedalgin-neo	Para600 + Caf100 + Cod base20 + Dipyrone300 + Phenobarbitol30	0
		Norman 1985	74	53	29	1.8 (1.0 - 3.3)	4.2 (2.2 - 48.5)
		Cater 1985	93	57	32	1.8 (1.1 - 2.9)	4.1 (2.3 - 19.8)
Solpadeine Max	Para1000 + Cod base20	0
Solpadeine Plus	Para1000 + Caf60 + Cod base13	1	61	29	4	6.2 (0.9 - 45.0)	4.2 (2.5 - 14.2)
Voltarol	Diclo25	4	502	53	15	3.6 (2.6 - 5.0)	2.6 (2.2 - 3.3)

Table 3 shows a sub-analysis of only those trials involving dental pain.

Drug	Constituents	Number of Trials	Number of Patients	Percent with Active	Percent with Control	RB (95% CI)	NNT (95% CI)
Anadin Extra	Para400 + Asa600 + Caf90	0
Askit	Asa530 + Caf110 + Aloxiprin140	0
Aspirine	Asa650 + Caf65	1	141	17	0	39.8 (2.4 - 648)	3.9 (2.7 - 6.7)
Aspro Clear	Asa1000	3	345	32	11	2.9 (1.8 - 4.8)	4.7 (3.4 - 7.6)
Codis	Asa1000 + Cod base 16	0
Cuprofen Plus	Ibu400 + Cod base 20	0
Disprin	Asa900	0
Disprin Extra	Asa600 + Para400	0
Feminax Ultra	Naprox500	5	402	62	7	8.9 (5.3 - 14.9)	1.8 (1.6 - 2.1)
Mersyndol	Para1000 + Cod base15 + Doxylamine succinate10	1	76	21	8	2.7 (0.8 - 9.3)	Not Calculated
Nurofen	Ibu400	49	5428	55	12	4.7 (4.2 - 5.2)	2.3 (2.2 - 2.4)
Panadeine 15	Para1000 + Cod base23	0
Panadol	Para1000	18	2171	40	9	4.4 (3.5 - 5.5)	3.3 (3.0 - 3.7)
Panadol Extra	Para1000 + Caf130	1	81	48	22	2.2 (1.1 - 4.2)	3.9 (2.2 - 18.0)
Paracodol	Para1000 + Cod base13	0
Paramol	Para1000 + Dihydrocodeine tartarate15	0
Pentalgin H	Naprox100 + Cod base8 + Caf50 + Dipyrone300 + Phenobarbitol15	0
Saridon	Para500 + Caf100 + Propifenazone300	1	301	23	2	9.2 (1.3 - 64.5)	4.9 (3.6 - 7.4)
Sedalgin-neo	Para600 + Caf100 + Cod base20 + Dipyrone300 + Phenobarbitol30	0
Solpadeine Max	Para1000 + Cod base20	0
Solpadeine Plus	Para1000 + Caf60 + Cod base13	1	61	29	4	6.2 (0.9 - 45.0)	4.2 (2.5 - 14.2)
Voltarol	Diclo25	3	398	51	11	4.6 (3.1 - 7.1)	2.5 (2.1 - 3.2)

To summarise the findings of our investigation we produced comparative figures (Figures 1 and 2 for all data and just dental studies, respectively) showing the NNTs and their 95% confidence intervals for each analgesic where calculable.

Figure 1: NNTs for all available data

Figure 2: NNTs for dental studies only

Comment

There are two main issues when looking at the evidence of acute pain efficacy of OTC analgesics. The first is the dearth of evidence in the public domain for some of these products. The second is what we are able to make of what evidence we have.

Dearth of evidence

Most of the OTC analgesics, including combination analgesics, were developed decades ago, as long ago as the 1950s, in times when trials were performed for registration purposes. Publication was infrequent. A good example is a review of 30 trials involving about 10,000 patients examining the analgesic efficacy of caffeine in combination with analgesics published in JAMA in 1984 [1]. Most of the data was unpublished then, and has remained unpublished subsequently. We know more about OTC drugs like paracetamol and ibuprofen from trials in which they have been used as active comparators than trials in which they themselves have been tested [2].

The dearth of evidence is not, therefore, surprising. It is, however, frustrating. For several OTC analgesics we have no reliable data, and for others the data available are inadequate - leading to very wide confidence intervals in Figures 1 and 2. This is a shame, because OTC analgesics, properly used, are effective for many people.

It is also the case that the case for analgesic combinations can be developed using evidence from closely related studies. A case in point is the combination of paracetamol and codeine, where relatively small amounts of information for some dose combinations is bolstered with evidence from other dose combinations [3].

What can we make of the evidence we have

The best evidence we have is from ibuprofen 400 mg (Nurofen), paracetamol 1000 mg (Panadol), naproxen 500 mg (Feminax Ultra), diclofenac 25 mg (Voltarol), and aspirin 1000 mg (Aspro), though the evidence is likely not to have come from testing of any particular product. All of these analgesics have usefully low NNTs in the range of about 2-4.

The evidence for combination analgesics is less clear, with predominantly no trials, or too few trials and patients available to make any judgement. This is a shame, because there is evidence elsewhere [3, for example] that combinations of analgesics can produce very good results.

Consumers can make up their own minds whether the expense of branded analgesics is worth it compared to the often much lower cost of unbranded - though that is a UK view, and certainly analgesics like paracetamol and ibuprofen are available in quantity and at low cost in the USA.

References

Laska EM, Sunshine A, Mueller F, Elvers WB, Siegel C, Rubin A. Caffeine as an analgesic adjuvant. JAMA 1984 251:1711-8.
Barden J, Derry S, McQuay HJ, Moore RA. Bias from industry trial funding? A framework, a suggested approach, and a negative result. Pain 2006 121:207-18.
Smith LA, Moore RA, McQuay HJ, Gavaghan D. Using evidence from different sources: an example using paracetamol 1000 mg plus codeine 60 mg. BMC Med Res Methodol. 2001 Jan 10;1:1.