Quality of osteoarthritis care for community-dwelling older adults

Authors


Abstract

Objective

To describe the quality of osteoarthritis care provided to community-dwelling elderly patients and to characterize arthritis-related function in these patients.

Methods

Two medical groups in the western United States participated in a practice-redesign intervention targeted at falls and mobility disorders, incontinence, and cognitive impairment, but not osteoarthritis. From 339 individuals reporting a diagnosis of arthritis, we collected information on demographics, functional status, and quality of care via patient interviews and mailed questionnaires. Eight quality indicators measured osteoarthritis care: 4 indicators measuring the provision of effective osteoarthritis care (osteoarthritis treatment indicators) and 4 measuring the provision of safe osteoarthritis care (medication safety indicators).

Results

The mean ± SD Short Form 12 physical component summary score was 35 ± 11 points, indicating a physically frail population. The overall indicator pass rate was 57.0% (95% confidence interval [95% CI] 53.9–60.2). Pass rates were higher for indicators of osteoarthritis treatment (63.5%, 95% CI 59.8–67.2) than for indicators of medication safety (43.8%, 95% CI 38.2–49.4). Patients with hip or knee pain had mean ± SD Western Ontario and McMaster Universities Osteoarthritis Index pain, stiffness, and function scores of 6.0 ± 4.0, 3.1 ± 1.7, and 25 ± 12 points, respectively.

Conclusion

Quality of osteoarthritis care for older adults is suboptimal, particularly with regard to medication safety. Given the high prevalence of osteoarthritis in older age groups, the population impact of any improvement in quality would be substantial. Quality improvement efforts for osteoarthritis should target appropriate use of and counseling regarding medications, as well as underuse of efficacious therapy for osteoarthritis.

INTRODUCTION

Osteoarthritis is a debilitating disease that affects 50–80% of the elderly population (1), and community-dwelling older adults report arthritis to be the leading cause of inability to perform an activity of daily living (2). Arthritis of the knee or hip in particular can severely compromise such activities as walking, climbing stairs, and using the bathroom (3). Treatments for osteoarthritis consist of patient education, exercise therapy, pharmacologic therapy, and surgical approaches (4). There has been little evaluation of the quality of care provided for persons with osteoarthritis, but the limited studies available are cause for concern. Previous data on quality of care for osteoarthritis suggest suboptimal care, with recommended care delivered to individuals with osteoarthritis ranging from 57% in the general population (5) to 31% among vulnerable elderly adults (6). However, these studies were limited by a small number of quality indicators and reliance on medical record review. Use of the medical record to measure quality may underestimate the quality of care for osteoarthritis due to lack of provider documentation regarding education and counseling, referral for physical modalities, or use of nonprescription medicines (7).

The present study used clinical data collected from older patients participating in an intervention to improve care for falls and mobility disorders, cognitive impairment, and urinary incontinence (8), and was an ancillary study to the main analysis (9). Data on health care quality for conditions other than those targeted for intervention were available from these patients. For patients with knee and/or hip osteoarthritis, disease-specific health status was assessed at the end of the study period using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) questionnaire (10). This study describes a community-dwelling sample of patients ≥75 years of age enrolled in the Assessing Care of Vulnerable Elders (ACOVE-2) intervention who also reported a diagnosis of osteoarthritis. For these patients, we provide a detailed assessment of performance on arthritis-related quality measures.

PATIENTS AND METHODS

Study sample.

The ACOVE-2 intervention, which was unrelated to osteoarthritis care, took place in 2 medical groups in the western United States. One medical group was a primary care group of 30 physicians caring for ∼20,000 patients (two-thirds managed care). The other group was a larger multispecialty group with 100 physicians caring for 140,000 patients (approximately half managed care). Patients older than age 65 were estimated to represent ∼50–60% of the patients being cared for (8).

Patients were eligible to participate in ACOVE-2 if they were ≥75 years of age, spoke English, had scheduled an appointment with a primary care practitioner, and screened positive for at least 1 of 3 target conditions: falls and mobility problems, cognitive impairment, or urinary incontinence. Practice personnel screened for these conditions via telephone approximately 1 week prior to a scheduled routine office visit. If the patient could not provide information, a proxy who knew the most about the patient's health was identified and was questioned. Eighteen percent of all respondents were surrogates.

A positive screen for falls and mobility disorders was defined as a history of ≥2 falls in the previous year, a fall with injury, or a fear of falling due to a gait or balance problem. Urinary incontinence generated a positive screen if it was bothersome enough that the patient would want to know more about how it could be treated. Cognitive impairment screening was positive if the patient recalled 0 or 1 of 3 items after a hiatus of at least 60 seconds, or if a surrogate reported that the patient recently had more trouble with memory for day-to-day happenings around the house. The institutional review boards at RAND; the University of California, Los Angeles (UCLA); and the Veterans Affairs Greater Los Angeles Health Care System approved the ACOVE-2 study and this additional study, which collected WOMAC scores.

Of the 3,010 patients age ≥75 years with a primary care physician appointment during the study period, 2,671 (89%) were screened for falls and mobility disorders, incontinence, and cognitive impairment. The remainder (n = 339) could not be reached/were non-English speaking (n = 211) or refused/had no surrogate decision maker (n = 128). Among the 2,671 screened individuals, 784 (29%) had a positive screen. Of these individuals, 649 (83%) consented to participate in the evaluation of the ACOVE-2 intervention; 5 of these 649 individuals had incomplete records. Of the 644 patients with complete medical record data, 566 (88%) completed interviews at the end of the study period. Of the remaining 78 individuals, 38 died during the study period and 40 refused to participate in the interview, were too ill, or could not be contacted.

For this analysis, we restricted the ACOVE-2 sample to those participants who answered “yes” to either of the following questions at the postintervention interview: “Has a doctor or nurse ever told you that you have arthritis?” or “Has a doctor or nurse ever told you that you have arthritis in your joints?” Of the 566 participants who took part in the quality of care interview and had usable data, 339 (60%) reported a physician or nurse diagnosis of arthritis.

ACOVE-2 intervention.

At one medical group, the ACOVE-2 intervention period took place from April 2002 to May 2003; at the other medical group, the intervention ran from July 2002 to August 2003. In the control practices during this time, when the patient came for the office visit, doctors were given the screening results without any further information. In the intervention practices, doctors received structured history/physical examination forms relevant to the conditions for which the patient screened positive to help guide the physician through an appropriate evaluation and management of the relevant condition. If patients screened positive for more than 1 condition, the intervention materials for one condition took priority in the following order: cognitive impairment, falls/mobility problems, and incontinence (8). These structured visit forms appeared on a patient's chart for only 1 visit during the intervention period.

Physicians in the intervention group also participated in a 3-hour educational program offering an efficient approach to managing each of the 3 target conditions, supplemented by condition-specific written educational materials. In addition, physicians in the intervention group had patient handouts available in the examination room for patient education on each of the target conditions.

Data.

This study analyzed data from the computer-assisted telephone interview that occurred at the end of the ACOVE-2 intervention. It was during this interview that participants reported whether they carried a diagnosis of arthritis, which then triggered research personnel to ask detailed questions regarding the duration and location of joint pain, as well as the questions used to measure quality of care for osteoarthritis. Patients were asked separately about the duration of knee pain (if applicable) and the duration of pain in all other joints taken together. At the same interview, patients were queried regarding their demographics and the number of prescription medications they were taking, and were also administered the Short Form 12 (SF-12) questionnaire, a measure of health-related quality of life with a mean ± SD general population score of 50 ± 10 (11). Details on the interview questions used to generate measures of quality of care can be found in Appendix A.

Quality of care for osteoarthritis was measured using quality indicators developed as part of the ACOVE project. These indicators were originally targeted at vulnerable elderly adults (those individuals age ≥65 years who were at increased risk for death or functional decline) (12) and were developed using a modified version of the RAND/UCLA appropriateness method (13). The ACOVE indicators aim to comprise the complete spectrum of care, including the domains of prevention, diagnosis, treatment, and followup. For the ACOVE-2 intervention, a clinical committee of geriatrics experts created a set of outpatient quality indictors by updating the original ACOVE indicators based on new medical literature, streamlining the set for the outpatient setting, and aiming the indicators toward patients age ≥75 years.

An example of a quality indicator is as follows: “IF a person age 75 or older is diagnosed with symptomatic osteoarthritis, THEN functional status and degree of pain should be assessed annually.” The “IF” statement determines eligibility for the care process in question, and the “THEN” statement specifies what care process should be performed. We used 8 quality of care indicators for patients with osteoarthritis (1). The indicators included 4 measures of osteoarthritis treatment: an annual assessment of an arthritis patient's functional status and degree of pain; exercise therapy for patients with newly diagnosed or prevalent osteoarthritis of the knee; education regarding the natural history, treatment, and self management of osteoarthritis; and referral to an orthopedic surgeon for patients with severe functionally compromising osteoarthritis of the knee or hip. The indicators also included 4 measures of medication safety in osteoarthritis care: acetaminophen as first-line pharmacologic therapy for osteoarthritis; advising patients treated with nonselective nonsteroidal antiinflammatory drugs (NSAIDs) of their risks; advising patients treated with cyclooxygenase 2 selective NSAIDs of their risks; and offering prophylaxis with a proton-pump inhibitor or misoprostol to patients treated with nonselective NSAIDs. Each quality indicator was scored on a pass/fail basis.

We present these quality indicator data in 2 ways. First, we calculated summary scores of quality of care for osteoarthritis for each patient. This indicates the percentage of indicators passed for a particular patient, who may have been eligible for anywhere from 0 to 8 indicators. We calculated the summary score as the total number of indicators passed divided by the total number of indicators for which the patient was eligible. Using the same approach, we calculated summary subscores for osteoarthritis treatment and medication safety. Second, we present the quality indicator data as pass rates in the study sample as a whole, where the numerator represents the number of indicators passed in the sample, and the denominator represents the number of indicators for which the study sample was eligible.

Within 1 month of the interview, patients who reported hip and/or knee pain during the interview were mailed the Likert-scale version of the WOMAC. Among these 238 individuals, 176 (74%) returned questionnaires with usable data. The WOMAC has been validated as a measure of disease-specific function for individuals with osteoarthritis of the hip or knee (10). It contains 5 questions about pain, 2 questions about stiffness, and 17 questions about function. Each item is scored on a 0–4 scale, with a higher score indicating more severe symptoms. All questions refer to symptoms or limitations occurring in the previous 48 hours. Examples of questions include: “How much pain have you had when going up or down stairs?” or “How much difficulty have you had getting in or out of a car, or getting on or off a bus?” Response options included “none,” “mild,” “moderate,” “severe,” or “extreme.”

Statistical analysis.

Statistical analyses were performed using SAS version 9.1.3 (SAS Institute, Cary, NC) and STATA version 9 (STATA, College Station, TX). For descriptive statistics, we report the mean ± SD for continuous variables and percentages for categorical variables. Because individuals' osteoarthritis quality of care summary scores and durations of joint pain were not normally distributed, we report these as medians with interquartile ranges (IQRs). Confidence intervals for proportions were adjusted to account for clustering of quality indicators within patients.

Because it was possible that the ACOVE-2 intervention affected quality of care for osteoarthritis, we compared the pass rates on individual quality indicators using chi-square tests; finding no evidence of significant differences between intervention and control groups, we report pooled results for quality.

Role of the funding sources.

The funding sources had no role in the design, conduct, or analysis of the study, or in the interpretation of the study findings.

RESULTS

The baseline demographic and clinical characteristics of the study sample are displayed in Table 1. Patients' mean age was 81 years, and 74% were women. The vast majority of study participants were white, approximately two-thirds had more than a high school education, and approximately half were married. The mean SF-12 mental component summary score was similar to the general population average, but the mean physical component summary score fell well below the general population average, reflecting the recruitment of a physically frail population. Patients received a mean of 6 prescription medications.

Table 1. Demographic characteristics and health status of the study sample (n = 339 except where noted)*
CharacteristicValue
  • *

    Values are the percentage unless otherwise indicated. SF-12 = Short Form 12.

  • See Patients and Methods section for definition of a positive screen.

  • Population mean ± SD for both of these scales is 50 ± 10.

Age at entry into study, mean ± SD years81.1 ± 4.9
Female sex74
Non-Hispanic white92
Married49
More than high school education (n = 338)64
Annual income ≥$35,000 (n = 319)39
Attends religious meetings at least weekly32
Screened positive for falls/mobility disorders79
Screened positive for cognitive impairment8
Screened positive for bothersome urinary incontinence40
SF-12 Physical Component Summary, mean ± SD score35.4 ± 11.4
SF-12 Mental Component Summary, mean ± SD51.1 ± 8.5
Mean ± SD number of prescription medications (n = 338)5.8 ± 3.2

The location of patients' joint pain is displayed in Table 2. Back, hip, and knee pain were each reported by approximately half of respondents. Patients with knee pain had a median pain duration of 5 years (IQR 3–15 years), whereas patients with pain in other joints had a median pain duration of 6 years (IQR 3–15 years).

Table 2. Location of joint pain*
LocationNo. (%) with pain
  • *

    Number of respondents varied from 336 to 339 (total sample = 339) for the items above. Responses are not mutually exclusive.

Back178 (52.8)
Knee174 (51.5)
Hip151 (44.8)
Finger142 (41.9)
Hand127 (37.5)
Shoulder118 (35.0)
Neck99 (29.5)
Wrist78 (23.1)
Ankle69 (20.5)
Ball of foot56 (16.7)
Toes56 (16.6)
Elbow40 (11.9)
Other joint31 (9.2)

When considered at the individual patient level, the median osteoarthritis quality of care score among 327 eligible patients was 50% (IQR 33–80%). The median osteoarthritis treatment subscore among 322 eligible patients was 67% (IQR 33–100%). The median medication safety score among 205 eligible patients was 50% (IQR 0–100%).

The pass rates for each quality indicator are shown in Table 3. Each indicator had a different set of eligible patients; for example, only patients with a diagnosis of arthritis who also reported joint pain were eligible for an annual assessment of pain and function (see Appendix A for details). The overall pass rate for the 8 measures was 57.0% (95% confidence interval [95% CI] 53.9–60.2). Pass rates were higher for osteoarthritis treatment indicators (63.5%, 95% CI 59.8–67.2) than for medication safety indicators (43.8%, 95% CI 38.2–49.4). Quality of care varied substantially among indicators, ranging from 27.4% for treating patients receiving nonselective NSAIDs with a proton-pump inhibitor or misoprostol to 72.7% for referral to an orthopedic surgeon of a patient with hip or knee osteoarthritis refractory to pharmacologic or nonpharmacologic therapy.

Table 3. Quality of osteoarthritis care*
IndicatorNo. of times eligibility metPass rate (%)
  • *

    NSAID = nonsteroidal antiinflammatory drug; COX-2 = cyclooxygenase 2.

IF a person age 75 or older is diagnosed with symptomatic osteoarthritis, THEN functional status and degree of pain should be assessed annually.26960.6
IF an ambulatory person age 75 or older has had a diagnosis of symptomatic osteoarthritis of the knee for >3 months and has no contraindications to exercise and is physically and mentally able to exercise, THEN a directed or supervised strengthening or aerobic exercise program should have been prescribed at least once.9144.0
IF an ambulatory person age 75 or older has had a diagnosis of symptomatic osteoarthritis for >6 months, THEN there should be evidence that education regarding the natural history, treatment, and self management of the disease was offered at least once.30068.7
IF a person age 75 or older with severe symptomatic osteoarthritis of the knee or hip has failed to respond to nonpharmacologic and pharmacologic therapy, THEN the patient should be offered referral to an orthopedic surgeon to be evaluated for total joint replacement within 6 months unless a contraindication to surgery is documented.11072.7
IF oral pharmacologic therapy is initiated to treat osteoarthritis, THEN acetaminophen should be the first drug used, unless there is a documented contraindication to use.13858.7
IF a person age 75 or older is treated with a nonselective NSAID, THEN the patient should be advised of the risks associated with the drug.10138.6
IF a person age 75 or older is treated with a COX-2 NSAID, THEN the patient should be advised of the risks associated with the drug.3450.0
IF a person age 75 or older is treated with a COX nonselective NSAID, THEN he or she should be offered concomitant treatment with either misoprostol or a proton-pump inhibitor.10627.4

For patients with hip or knee pain, mean ± SD WOMAC pain, stiffness, and function scores were 6.0 ± 4.0, 3.1 ± 1.7, and 25 ± 12 points, respectively. These scores are comparable with, if not slightly better than, those of a population of older adults with self-reported knee pain (14).

DISCUSSION

This study demonstrates the significant burden of osteoarthritis in a sample of community-dwelling older adults, and confirms previous reports that the quality of osteoarthritis care is suboptimal (5, 6). We report an overall 57% (95% CI 54–60) pass rate for osteoarthritis quality indicators, identical with the pass rate noted by McGlynn et al for randomly sampled adults of all ages (5) and higher than the 31% (95% CI 25–36) noted in the original ACOVE study for individuals age ≥65 years at increased risk of death or functional decline (6). Due to differences in the number and types of quality indicators used, sources of data (medical record versus interview), and patient populations assessed, it is not possible to make direct comparisons of quality of care across these 3 studies. Because our data derive solely from patient interview, this study avoids a potential methodologic limitation, namely, poor documentation as a potential source of spuriously low pass rates.

It is important to note that although the quality indicators used in ACOVE-2 measured care being delivered to the patient by the health care system as a whole, most patients in this study received their care from a generalist, with only 12% of the study sample seeing a rheumatologist at least once during the study period. Therefore, the performance noted here mostly reflects the actions of each patient's primary care physician or nurse practitioner and should be interpreted in this context.

This study also demonstrates that pass rates were noticeably lower for medication safety indicators (44%) than for indicators of osteoarthritis treatment (64%). This may reflect a more general trend of physicians to pay less attention to potential adverse effects of medications (15). Quality improvement efforts in osteoarthritis might best be targeted at ensuring safe medication use in this population, particularly given the substantial risk of adverse drug events in these patients who were taking a mean of 6 prescription medications (16).

For the subsample of patients with hip or knee pain, mean WOMAC scores of 6, 3, and 25 for pain, stiffness, and function, respectively, reflect moderate functional impairment, and are similar to, if not slightly better than, population norms for patients age ≥75 years with self-reported knee pain (14). By comparison, Canadian patients awaiting hip or knee arthroplasty had WOMAC pain, stiffness, and function scores of 12, 5, and 43, respectively (scores transformed to match scale used in this study) (17).

This study has important limitations. First, multiple factors affect the study's generalizability. The study was restricted to adults age ≥75 years who received care in only 2 medical groups in 1 state. Noticeable regional differences in care for older patients are well documented, even for such routine measures as influenza vaccination (18). Also, the patients in this study participated in a controlled intervention to improve care for falls/mobility disorders, incontinence, and cognitive impairment, which involved screening all participants (both in control and intervention groups) for the target conditions. It is possible that the process of identifying individuals with a history of falls and/or mobility disorders generated higher than usual rates of exercise therapy in both the control and intervention groups, because exercise therapy is often indicated in the treatment of falls. Although this could bias our results toward a higher than expected pass rate for exercise therapy, the observed performance rate of 44% among patients with knee osteoarthritis was nonetheless unacceptably low.

Second, the patient interview used to generate pass rates for quality indicators occurred at the end of the 13-month study period. It is possible that patients failed to recall details of the medical care they had received, which might lead to an underestimate of quality indicator pass rates. Conversely, patients might recall receiving a therapy that had not actually been administered, leading to a potential overestimate of pass rates. These issues pertain especially to surrogate respondents, who may not have been present with the patient when care was received, and patients with cognitive impairment, whose recall might be questionable.

Third, the patients in this study identified themselves as having arthritis, although not specifically osteoarthritis. However, given the high prevalence of osteoarthritis in this age group, it is unlikely that more than a few patients in the study sample had a different type of arthritis. Furthermore, all of the indicators are applicable to other types of arthritis with the exception of the indicator that assessed the use of acetaminophen as first-line drug treatment. This indicator would not be applicable to patients with rheumatoid arthritis. However, given that the population prevalence of rheumatoid arthritis is 2–3% in adults age ≥70 years (19), compared with a 60% prevalence of any type of arthritis in this study, we would estimate no more than 5% of patients in our sample to have rheumatoid arthritis, and hence the misclassification of rheumatoid arthritis would have a negligible effect on the scores we report.

Our data provide background and support for future quality improvement initiatives in osteoarthritis care for older adults. Given the high prevalence of osteoarthritis in this older age group, even small improvements in osteoarthritis care could have large and meaningful benefits for the older population as a whole.

Acknowledgements

The authors thank W. John Boscardin, PhD, Gerald F. Kominski, PhD, and the 2 anonymous peer reviewers for comments on a previous draft of the manuscript, and Rachel Louie, MS, for programming assistance. Robin P. Hertz, PhD, senior director of outcomes research/population studies at Pfizer Inc, provided valuable support. The authors also thank Patty Smith for administrative assistance with the ACOVE-2 study.

APPENDIX A

INTERVIEW QUESTIONS THAT ESTABLISHED ELIGIBILITY AND PASS RATES FOR THE QUALITY INDICATORS

The entire patient interview was completed using CATI (Computer Assisted Telephone Interview) technology. This allowed the interviewer's questions to be standardized but specific to the particular patient. For example, if a patient did not indicate any joint pain in the knee, then questions pertaining to knee osteoarthritis were automatically skipped.

Prior to the patient interview, patients were mailed a sheet on which they were to write all of their prescription medicines. If the patient had not filled out this sheet by the time of the interview, patients were asked to collect all of their actual medicines while the interviewer waited on the telephone. The patients then read off the medicines to the interviewer.

Below we list each quality indicator, indicate what was necessary for a patient to be eligible for the indicator, and then indicate how the indicator would be passed (or failed, when applicable).

IF a person age 75 or older is diagnosed with symptomatic osteoarthritis, THEN functional status and degree of pain should be assessed annually.

Eligibility criteria: Patients needed to state the presence of joint pain on a separate part of the questionnaire regarding pain, or answer yes to the following question: “Do you have pain in your joints that bothers you a lot on most days?”

Pass criteria: A “yes” response to either of the 2 questions below.

  • “In the past 12 months, has any doctor or nurse asked you about the severity of your joint pain?”

  • “In the past 12 months, has any doctor or nurse asked you if your joint pain affects the things you like or need to do?”

IF an ambulatory person age 75 or older has had a diagnosis of symptomatic osteoarthritis of the knee for >3 months and has no contraindications to exercise and is physically and mentally able to exercise, THEN a directed or supervised strengthening or aerobic exercise program should have been prescribed at least once.

Eligibility criteria:

  • Knee pain that had lasted at least 3 months.

  • Health did not limit the patient's ability to bathe or dress.

  • Patients or surrogates indicated that the patient did not have severe dementia.

Pass criterion: A “yes” response to the question below. “You told me earlier that you have had knee pain for >3 months. Have you ever been sent to physical therapy or an exercise group or class for your knee pain?”

IF an ambulatory person age 75 or older has had a diagnosis of symptomatic osteoarthritis for >6 months, THEN there should be evidence that education regarding the natural history, treatment, and self management of the disease was offered at least once.

Eligibility criteria: Pain in any joint lasting at least 6 months.

Pass criteria: A “yes” response to at least 1 of the 3 bullets below.

“Has any doctor or nurse ever talked to you about”:

  • What your arthritis or joint pain will be like as time goes on, or the natural history of arthritis?

  • How to keep your arthritis or joint pain from getting worse?

  • How your arthritis can be treated?

IF a person age 75 or older with severe symptomatic osteoarthritis of the knee or hip has failed to respond to nonpharmacologic and pharmacologic therapy, THEN the patient should be offered referral to an orthopedic surgeon to be evaluated for total joint replacement within 6 months unless a contraindication to surgery is documented.

Eligibility criteria:

  • Joint pain in the knee or hip, AND

  • Answer “yes” to the following question: “Does your arthritis or joint pain stop you from performing daily activities, like standing or walking?”

Pass criteria: A “yes” response to either or both of the 2 questions below.

  • “Did any doctor or nurse say that you should have surgery or joint replacement?”

  • “Did any doctor or nurse recommend you see an orthopedic surgeon?”

Fail criteria:

  • A “no” response to both questions above, AND

  • The pain that stopped the patient from standing or walking had been going on for more than 6 months.

IF oral pharmacologic therapy is initiated to treat osteoarthritis, THEN acetaminophen should be the first drug used, unless there is a documented contraindication to use.

Eligibility criteria: A “yes” response to following question: “Has any doctor or nurse recommended any treatments for your joint pain?”

Pass criteria: A “yes” response to the question below.

“Some doctor or nurses suggest trying Tylenol or acetaminophen as the first medication for arthritis pain. Did any doctor or nurse recommend you try Tylenol before other medications for your joint pain?”

IF a person age 75 or older is treated with a nonselective NSAID, THEN the patient should be advised of the risks associated with the drug.

Eligibility criteria: Nonselective NSAID on medication list, or a “yes” response to both of the following 2 questions.

  • “Do you take any nonprescription medications like ibuprofen, naproxen, Advil, Motrin, Aleve, or Naprosyn? These nonprescription medications are known as nonsteroidal antiinflammatory drugs or NSAIDS.”

  • “Did a doctor or nurse recommend taking this medication?”

Pass criteria: A “yes” response to the question below.

“Did any doctor or nurse tell you about the possible side effects or risks from taking this medication?”

IF a person age 75 or older is treated with a COX-2 NSAID, THEN the patient should be advised of the risks associated with the drug.

Eligibility criteria: COX-2 NSAID on medication list.

Pass criteria: A “yes” response to the question below.

“Did any doctor or nurse tell you about the possible side effects or risks from [COX-2 NSAID]?”

IF a person age 75 or older is treated with a COX nonselective NSAID, THEN he or she should be offered concomitant treatment with either misoprostol or a proton-pump inhibitor.

Eligibility criteria: Nonselective NSAID on medication list, or a “yes” response to both of the following 2 questions.

  • “Do you take any nonprescription medications like ibuprofen, naproxen, Advil, Motrin, Aleve, or Naprosyn? These nonprescription medications are known as nonsteroidal antiinflammatory drugs or NSAIDS.”

  • “Did a doctor or nurse recommend taking this medication?”

Pass criteria: Proton-pump inhibitor or misoprostol on patient's medication list, or a “yes” response to the question below.

“Did any doctor or nurse recommend you take another medication to protect your stomach or intestines or from developing an ulcer?”

(If the patient wasn't sure, the patient was prompted as follows: “Medications such as misoprostol or proton-pump inhibitors such as Aciphex, Nexium, Prilosec, Prevacid, or Protonix?”)

Ancillary