Awareness and use of over-the-counter pain medications: a survey of emergency department patients.

ABSTRACT

Background. A survey study was done to determine basic knowledge about and use of over-the-counter (OTC) pain medications among patients seen in our emergency department (ED). The hypothesis was that knowledge about OTC pain medications would be correlated with demographic characteristics.

Methods. The study was a prospective 21-question survey during randomized 4-hour shifts over a 6-week study period.

Results. Of the 213 subjects who participated, 143 (67%) reported having used some form of nonsteroidal anti-inflammatory drug (NSAID), and 127 (60%) had used acetaminophen products recently. Women were more aware of toxic interactions and gastrointestinal (GI) irritation related to these medications. Ibuprofen use correlated with age (younger individuals used more) and having no primary physician (those without a physician used less). Knowledge about GI effects correlated with age, sex, and education. Knowledge about renal and hepatic problems correlated with age and education.

Conclusion. Many patients use OTC pain medications. Many are ill informed about their use and side effects. More education is needed.

**********

OVER-THE-COUNTER (OTC) pain medications are commonly used in both adult and pediatric patient populations for analgesic effects and in children for fever. (1-6) Adverse effects of their use include GI distress, allergic reactions, kidney failure and liver toxicity with prolonged use and overdoses. (1,2,4,5,7-14)

Although all Food and Drug Administration (FDA)-approved OTC pain medications are appropriately labeled with dosage information, not all brands supply the consumer with detailed information concerning adverse effects and contraindications. Indiscriminate use of OTC medications such as pain medications is a concern to the medical community.

Demographic factors such as age, education, and access to medical care may influence patients' awareness of the interactions and side effects of OTC pain medications. The purpose of the present survey study was to determine predictors of patients' knowledge about OTC pain medications and their side effects. The hypothesis was that a patient's knowledge and use of OTC pain medications are correlated with demographic characteristics.

METHODS

Study Design

We prospectively surveyed ED patients seen during randomized 4-hour shifts. The study population consisted of all ambulatory adult patients who came to the triage desk of our facility, an urban ED with a residency program. Surveys were conducted over a 6-week period in randomized 4-hour shifts on both weekdays and weekends to ensure an appropriate sampling of all ED patients. Shifts were randomized to include equal representation for all times of the day and days of the week. Participants included men and women aged 18 years and older who were able to answer questions. Individuals who were unable to answer questions because of psychiatric, medical, or traumatic illness or a language barrier were excluded from participation in the study. They were also excluded if they refused to participate. A log was kept of demographic variables of those who were considered able to participate but requested not to be included. The researchers reminded patients who were taking NSAIDs or acetaminophen by prescription that an y statements in the survey were not to take precedence over advice or instructions given by their physician or pharmacist. The study was approved as exempted by the Human Subjects Review Committee at this institution.

Study Protocol

Patients aged 18 and older, without evidence of alcohol intoxication, drug use, or psychiatric or severe illness were approached at the triage area. Two research assistants trained over a 4- hour shift approached patients for verbal consent to participate in the study. If they agreed, a 21-question written survey regarding knowledge, about ibuprofen, naproxen, aspirin, and acetaminophen was administered (Appendix). The survey was self-administered, with the assistance of the researchers if needed. Responses to the survey were recorded.

Demographic information included age, sex, ethnicity, marital status, education level, insurance status, whether they have a primary care physician, and personal use of OTC pain relievers. Brand and generic names were both used to avoid confusion.

Next, health information was obtained including whether the subjects discussed use of OTC pain medications with their doctors, whether they regularly consumed alcohol, whether they had applicable health problems, and whether they had had any side effects from OTC pain medications. Applicable health problems were defined as asthma, renal disease, and liver disease.

Five questions about OTC pain medications were then asked regarding knowledge of toxic interactions with other medications and the association of OTC pain relievers with GI irritation, asthma, kidney toxicity, and liver damage. We concluded the survey by asking participants if the awareness of side effects obtained by completing this survey would cause a behavior change in their use of OTC pain medications. The answer was dichotomized as yes or no.

Key Outcome Measures

Key outcome measures were differences in knowledge base according to demographic characteristics.

Analytical Methods

Results were tabulated and analyzed using the SPSS software program (SPSS version 9.0, Chicago, Ill). Demographic variables were analyzed using a Pearson's chi-square for categorical data. Subgroup analysis was based on applicable health problems, insurance status, and sex. Odds ratios and their confidence intervals were calculated for ail significant differences.

A multivariable logistic regression analysis was done correlating the use/nonuse of NSAIDs and the results of 5 knowledge questions (interactions, GI distress, asthma, kidney disease, and liver problems) as outcome variables. Statistics (estimates and 95% confidence intervals [CI]) are reported for significant predictors only. Factors included in the models include age, sex, race, marital status, education, and whether the patient had a primary physician. A P value of < .05 was considered statistically significant.

RESULTS

Demographics

During 42 randomized 4-hour shifts, 232 subjects were approached. Nineteen patients (8%) who were considered eligible to participate refused. Thus, 213 subjects participated over the 6-week study period during randomized 4-hour shifts. Mean age of participants was 36 [+ or -] 13.7. There were 123 women (58%) and 89 men (42%). One person chose not to identify his or her sex. Of the 213 participants, 107 (50%) were white, 60 (28%) were African-American, 24 (11%) were Hispanic, 14 (7%) were Asian, and 8 (4%) were other races. Sixty-nine patients (32.4%) had private insurance, 50 (23.5%) had state insurance, and 94 (44.1%) had no insurance. Seventy patients (33%) went to the ED with a trauma complaint and 143 (67%) with a medical complaint. Table i summarizes the study participants.

There was significant history of use of OTC pain medications in our patient population. Nonsteroidal anti-inflammatory drugs were used by 143 (67%) of the subjects and acetaminophen by 127 (60%). Breakdown showed that 108 patients used ibuprofen, 22 used naproxen, and 62 used aspirin.

Health Information

In 155 (73%) of cases, subjects had never discussed the use of OTC pain medications with doctors. Thirty subjects (14%) admitted to regularly consuming alcohol. Applicable health problems in the 213 participants were asthma in 17, kidney problems in 8, and liver problems in 8. Ten patients (5%) were currently taking Tylenol (acetaminophen), 4 (2%) Tylenol with codeine, 15 (7%) vicodin, 24 (11%) ibuprofen, 4 (2%) naproxen, and 8 (4%) aspirin products. Adverse effects due to the use of NSAIDs were seen in 24 (11%); those due to acetaminophen were seen in 6 (3%).

Knowledge

Table 2 shows responses to questions concerning knowledge of the adverse effects and interactions of OTC pain medications. A majority of patients knew about GI distress in relation to use of NSAIDs; however, fewer knew about adverse interactions and associations with asthma, liver disease, and kidney problems.

Behavioral Change

When asked whether they would change their behavior on the basis of presently gained awareness about OTC pain medications, 37 (17.4%) answered no and 176 (82.6%) answered yes.

Subgroup Analysis

Of the 17 asthmatics, 4 (24%) were aware of the link between aspirin and asthma attacks. Of the 8 with kidney problems and 8 with liver problems, 4 of each or 50% were aware of the link between kidney and liver toxicity with medication use. These numbers were too small for statistical analysis.

Patients with insurance were more likely to have a primary doctor (P < .001, OR = 16.6; 95% CI = 7.8-35.4). There was no difference between those with and those without insurance in awareness of toxic interactions (P = .78), GI distress (P = .57), asthma (P = .78), or liver effects (P = .89). Having insurance accounted for a significantly greater difference in awareness about kidney effects (P= .04, OR = 1.9; 95% CI = 1.03-3.4). Only 10 (11%) of the 94 uninsured people in our study had a primary care physician, as opposed to 79 (72%) of the 109 with insurance (OR = 16.6, 95% CI = 7.8-35.4)

By chi-square analysis, women were more likely to use ibuprofen than men (72 vs 36 [58% vs 40%], P = .013). However, there was no difference between sexes in use of naproxen (P = .31), aspirin (P = .36), or acetaminophen (P = .22). More women were aware of potential toxic interactions with OTC medication use (OR 1.6, 95% CI 0.9-2.7). More women had a primary care physician than men (P = .005). More women stated they would discuss NSAID use with their own doctors (P = .013) and more men (22 men vs 7 women) admitted to using alcohol (P K .001). Women were more aware of GI distress in association with NSAID use (P = .01, OR = 1.8, 95% CI = 1.02-3.1), but there was no difference in awareness of asthma, kidney problems, or liver problems (P> .05). Women were more likely to change their behavior after gaining this awareness about OTC pain medications (P= .006, OR = 2.7; 95% CI = 1.3-5.6). Odds ratios comparing men and women on the knowledge questions and the behavior change question are summarized in Table 2.

Regression Analysis 

Regression analysis of the outcome variable of use/nonuse of ibuprofen was positive for age (young used more, P = .046) and absence of a personal physician (those without a physician were less likely to use ibuprofen, P = .04). The median age was 30 (IQ range = 23) for ibuprofen users and 38 (IQ range = 20) for nonusers. Subjects without a physician were about half as likely to be taking ibuprofen.

Neither the awareness of interactions with other medications nor awareness of the relationship with asthma was predicted by any of the factors. Awareness about GI distress with OTC pain medications was predicted by age (young less aware, P < .001), sex (men less aware, P = .01), and education (less educated less aware, P = .01). Awareness of renal problems associated with OTC pain medications was predicted by age (young less aware, P = .01) and education (less educated less aware, P = .03). Awareness of hepatic problems and OTC pain medications was predicted by age (young less aware, P = .03) and education (less educated less aware, P = .04).

Regression analysis for the significant predictors of age, sex, and education are summarized in Table 3. Race and marital status were not predictors of any of the outcome variables.

DISCUSSION

Over-the-counter NSAIDs and acetaminophen are among the most commonly used medications. Publications regarding side effects are widely available to health care personnel, but this information may not be readily accessible or comprehensible to the public. The FDA requires OTC pain medications to bear warning labels. These warning labels often advise consumers to consult their physician for additional information. However, finding a doctor is a challenge for those without medical insurance. We found that the insured in our study were more than 16 times more likely to have a personal physician than uninsured patients.

Sales of OTC analgesics have been growing. (15,16) Obtaining OTC pain medications is much cheaper and easier than seeing a doctor and going through the process of obtaining a prescription drug. As a result, there is a high potential for inappropriate use and side effects, especially if people are unaware of the risks. We found significant gaps in our patients' knowledge about OTC pain medications. More than 40% did not know about toxic interactions or GI side effects, more than 60% did not know about the relationship between these drugs and hepatic and renal disease, and more than 80% did not know about the relationship between aspirin and adverse effects in asthmatics.

That most of our patients were aware of GI side effects was not surprising, since the adverse effect most often recalled was NSAID-associated GI distress, including nausea, stomach upset, and ulcers. The low level of knowledge about OTC pain medications' potential adverse effects on the kidney and liver and on asthmatics was alarming. There is obviously a great need for patient education about OTC pain medications.

Groups that tended to be better informed about the adverse side effects in general were patients with more education and the elderly. In general, elderly and more educated patients knew more about NSAIDs and took less of them. The literature indicates that the elderly are among those at highest risk for renal and GI side effects of NSAIDs. (2,4,17) Older patients' increased awareness of risks in our study may reflect consciousness of their own susceptibility to drug side effects. Physicians may also have played a role in educating them, since virtually all elderly Americans have Medicare coverage and thus access to medical services.

Patients with more education are also among the groups that are better informed. They may have greater ability to understand warnings, better access to information on OTC analgesics, or more self-motivation to learn about side effects. In addition, as with the elderly, they may be more likely to have a personal physician.

Women were better informed specifically about GI side effects, and they also were more likely to have GI side effects. Female patients reported being more likely to ask doctors about OTC pain relievers and more likely to see a doctor. However, they were not more knowledgeable about toxic interactions, renal problems, and liver problems.

One of our initial hypotheses was that individuals without a primary care physician would be less aware of side effects than those with a primary care physician. We did find that those subjects without a primary care physician were less likely to be taking ibuprofen, but we did not find the presence of a personal physician to be a predictor of awareness on any of the knowledge questions. Interestingly, of those with a primary care doctor, less than 50% reported speaking to a physician about use of OTC pain relievers. Only 16% of patients who did not have a primary care doctor reported having ever spoken to a physician about OTC pain reliever use. Our findings suggest that, in the interest of prevention, physicians in primary care and in the ED need to increase efforts in educating patients in general.

In light of the risks of OTC pain medications, a great majority (83%) of all patients in the sample indicated they would alter their use of these drugs. Most stated that they would use them more cautiously. Thus, a majority of patients found this information useful.

All consumers of OTC pain reliever medicines should be encouraged to take a more active role in the maintenance of their own health. Physicians and pharmacists should take the time to counsel patients about adverse effects of OTC drugs. Consumers of OTC pain medications must be urged to read product ingredient labels, since prescription and nonprescription cold medications can contain alcohol, acetaminophen, or NSAIDs. They should be reminded not to exceed the recommended daily dosages, since doing so will not provide more rapid relief. Public service announcements can also be beneficial in increasing awareness.

Reliable written and/or audible information in nontechnical language should be made easily accessible and noticeable at places wherever OTC pain relievers are sold. The FDA has made a considerable effort to increase public awareness. In 1999, the FDA launched a public education campaign to educate consumers about OTC drugs and to help them read labels. Key aspects of the OTC label include (1) headings and subheadings, boldface type, bullets, and simpler concise language; (2) alphabetic listing of active ingredients; (3) a list of uses to help consumers compare products and find the appropriate medicine; and (4) placing all warnings in a continuous space. (18) Beginning March 1999, OTC drug manufacturers have about 2 years to comply with the new labeling, so it is too early yet to tell whether these mandated changes have enabled consumers to be better informed. In October 1999, the FDA and the National Association of Chain Drug Stores launched a joint public awareness campaign for women on safe medicine use c alled "Take Time to Care." In 14 cities nationwide, free brochures encouraging proper use of medications were distributed in 13 languages at clinics, conferences, universities, and community health centers. These brochures were also made available on the FDA website, and a toll-free hotline (1877 RX WOMEN) was established. Women were targeted for this public education effort because of their role in health care decisionmaking for the family and their more frequent use of medication. (19) There is potentially much to gain in these efforts to better inform patients. With education, patients can take an active role in preventing adverse drug reactions and longterm side effects.

LIMITATIONS AND FUTURE QUESTIONS

In our survey, we did not provide a comprehensive list of side effects to query the patient. Rather, we chose a-few major side effects to ask about, and the question provided an example of which group (s) was at risk. There may be other relevant risks that patients did not know about and risks that they did know about but were not asked. In addition, we could not verify patient-reported health problems in their charts. Patients may have withheld medical information or reported undiagnosed health problems. Another limitation is that we could not assess whether patients experienced asymptomatic side effects. There may also be a recall bias in which a side effect that occurred long ago is forgotten and not reported. As a result, the number of patients who had side effects could be higher than what we have found.

We did not study teenagers. Other studies have shown that teenagers underestimate the hazards of OTC overdose. (20 21) According to one study, many adolescents attempt suicide through overdose in search of attention but not truly wishing to die. (20) Lethal hepatotoxicity due to such behavior can perhaps be prevented if young people can appreciate the dangers of acetaminophen. It would be interesting to see the responses from this age group to a survey such as ours.

Despite our study's limitations, we believe it shows that a significant number of ambulatory patients presenting themselves at ED triage are ill informed about the risks of OTC drugs. Future studies should assess the prevalence of regular patients seen at ED triage who are currently misusing OTC pain relievers, examine the efficacy of the "Take Time to Care" campaign, and evaluate whether FDA-mandated changes in labeling have allowed consumers to better understand labels.

TABLE 1.

Demographics of Study Participants


Age
 Mean (SD)                36 (13.7)
Sex
 Male                     89 (42%)
 Female                  123 (58%)
Race
 White                   107 (50%)
 Black                    60 (28%)
 Hispanic                 24 (11%)
 Asian                    14 (7%)
 Other                     8 (4%)
Education
 Less than high school    35 (16.5%)
 High school              75 (35%)
 Some college            102 (48%)
 Unknown                   1 (0.5%)
Insurance
 None                    94 (44.1%)
 Public                  50 (23.5%)
 Private                 69 (32.4%)
Primary care physician?
 Yes                      89 (42%)
 No                      124 (58%)
Table 2

Knowledge About Interactions and Adverse Effects of Over-the-Counter
Pain Medications

                                          Sex differences
                                 No. (%)  OR (95 % CI)

Toxic interactions
    Yes                         124 (58)  1.6 (0.9-2.7)
    No                           89 (42)
Gastrointestinal irritation
 with NSAIDs
    Yes                         130 (61)  1.8 (1.02-3.1)
    No                           83 (40)
Asthma associated with aspirin
    Yes                          38 (18)  NS
    No                          175 (82)
Kidney toxicity with NSAIDs
    Yes                          68 (32)  NS
    No                          145 (68)
Liver toxicity with
 acetaminophen
    Yes                          76 (36)  NS
    No                          137 (64)
Change in behavior?
    Yes                         176 (82)  2.7 (1.3-5.6) *
    No                           37 (17)

* P < .01.

OR = Odds ratio, NSAIDs = nonsteroidal anti-inflammatory drugs, NS = not
significant.
TABLE 3.

Regression Analysis for Predictors of Ibuprofen Use and Knowledge About
Adverse Effects of Nonsteroidal Anti-inflammatory Drugs

                                   Age              Sex
                           P Value     R       P Value

Ibuprofen use               .046     -0.08       NS
Knowledge about drug         NS                  NS
 interactions
Knowledge about             <.001    0.27       .01
 gastrointestinal effects
Knowledge about              NS                  NS
 asthma
Knowledge about              .01     0.13        NS
 renal problems
Knowledge about              0.3     0.10        NS
 hepatic problems

                                  Sex          Education
                             OR (95 % CI)    P Value

Ibuprofen use                                  NS
Knowledge about drug                           NS
 interactions
Knowledge about            0.39 (0.20-0.80)   .01
 gastrointestinal effects
Knowledge about                                NS
 asthma
Knowledge about                               .03
 renal problems
Knowledge about                               .04
 hepatic problems

                                 Education          Personal
                                                   Physician
                              OR (95 % CI) *     P Value

Ibuprofen use                                     .04
Knowledge about drug                               NS
 interactions
Knowledge about            0.26 (0.10-0.65) (1)    NS
 gastrointestinal effects  0.47 (0.23-0.96) (2)
Knowledge about                                    NS
 asthma
Knowledge about            0.27 (0.09-0.77) (1)    NS
 renal problems            0.60 (0.30-1.20) (2)
Knowledge about            0.27 (0.10-0.75) (1)    NS
 hepatic problems          0.69 (0.36-1.36) (2)

                           Personal Physician
                            OR (95 % CI) *

Ibuprofen use              0.47 (0.23-0.98)
Knowledge about drug
 interactions
Knowledge about
 gastrointestinal effects
Knowledge about
 asthma
Knowledge about
 renal problems
Knowledge about
 hepatic problems

* Comparison to college level or higher. ((1) Less than high school
education. (2) High school education.)

Correlation coefficient is given for age as a continuous variable. Odds
ratios (OR) and 95% confidence intervals (CI) are given for sex,
education, and personal physician as categorical variables. Only
significant correlations or odds ratios are included.

References

(1.) Bjorkman DJ: Current status of NSAID use in the US: risk factors and frequency of complications. Am J Med 1999; 107: 3S-8S

(2.) Silagy CA, McNeil JJ, Donnan GA, et al: Adverse effects of low-dose aspirin in a healthy elderly population. Clin Pharmacol Ther 1993;54:84-89

(3.) McGoldrick MD, Bailie CR: Nonnarcotic analgesics: prevalence and estimated economic impact of toxicities. Ann Pharmacother 1997;31:221-226

(4.) Carson JL, Strom BL: Nonsteroidal antiinflammatory drugs. Principles of Geriatric Medicine and Gerontology. Hazzard WR, Biermal EL, Ettinger WI-I, et al (eds). New York, McGraw Hill, 3rd Ed, 1994

(5.) McLaughlin JK, Lipworth L, Chow W, et al: Analgesic use and chronic renal failure: a critical review of the epidemiologic literature. Kidney Int 1998;54:679-686

(6.) Barrett N: Parental knowledge of different acetaminophen concentrations for infants and children. Acad Emerg Med 2000;7:718-721

(7.) Bray GP, Harrison PM, O'Grady JG, et al: Long-term anticonvulsant therapy worsens outcome in paracetamol-induced fulminant hepatic failure. Hum Exp Toxicol 1992;11:265-270

(8.) Fry SW, Seeff LB: Hepatotoxicity of analgesics and anti-inflammatory agents. Gastroenterol Clin North Am 1995;24:875-905

(9.) Miners JO, Attwood J, Birkett DJ: Determinants of acetaminophen metabolism: effect of inducers and inhibitors of drug metabolism on acetaminophen's metabolic pathways. Clin Pharmacol Ther 1984;35:480-486

(10.) Miners JO: Drug interactions involving aspirin (acetylsalicyclic acid) and salicylic acid. Clin Pharmacokinet 1989;17:327-344

(11.) Zimmerman HJ, Maddrey WC: Acetaminophen (paracetamol) hepatoxicity with regular intake of alcohol: analysis of instances of therapeutic misadventure. Hepatology 1995;22:767-773

(12.) Schiodt FV, Rochling FA, Casey DL, et al: Acetaminophen toxicity in an urban county hospital. N Engl J Med 1997;337:1112-1117

(13.) Hedman J, Kaprio J, Poussa T, et al: Prevalence of asthma, aspirin intolerance, nasal polyposis and chronic obstructive pulmonary disease in a population-based study. Int Epidemiol Assoc 1999;28:717-722

(14.) Whitcomb DC, Block GD: Association of acetaminophen hepatotoxicity with fasting and ethanol use. JAMA 1994;272:1845-1850

(15.) Nonprescription NSAIDS. efficacy and safety. Available at: http://pharminfo.com/pubs/msb/nsaids.html. Accessed October 10, 2000

(16.) OTC Retail Sales. Available at: http://www.ndmaininfo.org/facts/factsRetail.html. Accessed October 1, 2000

(17.) Cooper JW: Adverse drug reaction-related hospitalizations of nursing facility patients: a 4-year study. South Med J 1999;92:485-490

(18.) Farley D: Label Literacy for OTC drugs. FDA consumer (online). Available at: http://www.fda.gov/fdac/feattires/1997/497_otc.html. Accessed October 10, 2000

(19.) Women's Health: Take time to care. Available at: http://www.fda.gov/womens/taketimetocare/hilights.html. Accessed October 10, 2000

(20.) Harris HE, Myers WC: Adolescents misperceptions about the dangerousness of acetaminophen in overdose (Abstract). Suicide Life Threat Behav 1997;27:274-277

(21.) Huott MA, Storrow AB: A survey of adolescents' knowledge regarding toxicity of OTC pain medications. Acad Emerg Med 1997;4:214-218

RELATED ARTICLE: KEY POINTS

* Many patients use over-the-counter (OTC) pain medications.

* Many are unaware about how to use these medications and may use them improperly.

* Patients are often unaware of adverse effects of OTC pain medications.

* Physicians should take the opportunity to educate patients about the use and side effects of OTC pain medication.

From the Department of Medicine, Division of Emergency Medicine, University of California, Davis.

Reprint requests to Amy A. Ernst, MD, University of California, Davis, Department of Medicine, Division of Emergency Medicine, 2315 Stockton Blvd, Sacramento, CA 95817.