Secondary Logo

Journal Logo

Ethical Issues

“We Don’t Carry That”—Failure of Pharmacies in Predominantly Nonwhite Neighborhoods to Stock Opioid Analgesics

  • Free


Hertzberg Palliative Care Institute, Department of Geriatrics and Adult Development, and Department of Biomathematical Sciences, Mount Sinai School of Medicine, New York, New York

N. Engl. J. Med., 342: 1023–1026, 2000

Based on the authors’ observation that many black and Hispanic patients who received palliative care at their hospital could not obtain prescribed opioids from their neighborhood pharmacies, they investigated the availability of commonly prescribed opioids in New York pharmacies. They surveyed a randomly selected sample of 30% of New York City pharmacies to obtain information about their stock of opioids. For each pharmacy, United States census estimates for 1997 were used to ascertain the racial and ethnic composition of the neighborhood (defined as the area within a 0.4 km [0.25 mile] radius of the pharmacy) and the proportion of residents who were older than 65 yr of age. Data on robberies, burglaries, and arrests involving illicit drugs in 1997 were obtained from the precinct in which each pharmacy was located. A generalized linear model was used to examine the relationship between the racial or ethnic composition of neighborhoods and the opioid supplies of pharmacies, while controlling for the proportion of elderly persons at the census-block level and for crime rates at the precinct level. Pharmacists who represented 347 of 431 eligible pharmacies (81%) responded to the survey. A total of 178 pharmacies (61%) did not have enough opioids to treat patients with severe pain. Only 25% of pharmacies in predominantly nonwhite neighborhoods (those in which less than 40% of residents were white) had opioid supplies that were adequate to treat patients in severe pain, compared with 72% of pharmacies in predominantly white neighborhoods (those in which at least 80% of residents were white). It was concluded that pharmacies in predominantly nonwhite neighborhoods of New York City do not stock enough medications to adequately treat patients who are in severe pain.


The article reports the results of a study that found that the availability of opioid analgesics in community pharmacies appears reduced in minority neighborhoods in New York City. The data were assessed with information on race, household income, educational level, proportion of persons over age 65 yr, and information on crime in the area. The results of the study are interesting because they provide insight into how demographic and medicolegal variables affect opioid access.

First, the authors found in the study that roughly half of the community pharmacies did not have adequate stock as defined by the authors to treat a patient in severe pain. Nonwhite race was significantly associated with this finding. Although it may be tempting to somehow attribute this limitation to racial discrimination, which indeed, the accompanying editorial attempts to do, what is of importance in attempting to understand this result is what information is not stated in the article. The contribution of factors of income, educational level, and proportion of persons over age 65 yr to the level of opioids in the pharmacies was not reported. The influence of these factors would surely provide a more complete picture of the reasons for potential reductions in opioid stocking and access in community pharmacies. In addition, attention to cultural norms is not provided in the study; for example, there is a relatively low availability of opioids in “Asian” neighborhoods (apparently combining Chinese, Japanese, Korean, Filipino, Vietnamese, and other persons into this one group); yet some individuals within these groups are averse to pain medications owing to tenets of their cultural philosophy. Attention to cultural differences may explain at least part of the differential between minority and nonminority neighborhoods; this represents an area that should have been explored by the authors.

The study reports that the majority reason for not stocking opioids is that the demand for the product is limited. Practically speaking, pharmacies do not stock these items because they do not obtain the necessary business to sell these products. Thus, the rational pharmacy would not purchase opioids in what the authors deem appropriate levels. This reason would seem to reflect practical business realities rather than reaching for causes based primarily in racial discrimination.

In addition, it is also interesting to note that the legal system seems to drive entity behavior in this area as it does in others. 1,2 The second most frequently cited reason for not stocking opioids is the high administrative costs associated with working with opioid agents—the significant state and federal legal requirements that represent high transaction costs to community pharmacies. These costs may not be easily or effectively shouldered by small business pharmacies, particularly in the context of relatively low demand. Economic theory would suggest that rational pharmacies would substitute away from such high cost, low demand products 3; again, this would not seem to be a racially motivated decision.

One consideration that the authors do not address is why the authors’ medical center itself did not dispense the opioids to the patients instead of sending patients to community pharmacies. The authors indicated at the outset that the rationale for the study was their observation that patients receiving palliative care at their teaching hospital could not obtain opioids from community pharmacies. It would seem that medical centers would be the logical and primary site to obtain prescription products, and a discussion of the limitations of using the medical center to provide desired opioids would have been helpful. Indeed, it would seem that clinic visits linked to opioid dispensing would be a solution to the problem that might alleviate the problems of patients accessing these drugs in their community as well as provide important opportunities for providers to assess their patients’ clinical conditions and an opportunity to address their patients’ clinical and social concerns.

What is of great concern is the editorial that accompanied the article. In the editorial, the authors right from the start mischaracterize the paper’s conclusion by stating that “In this issue of the Journal, Morrison et al. show that black and Hispanic [sic] patients with severe pain are less likely than white patients to be able to obtain commonly prescribed pain medicines, because pharmacies in predominantly nonwhite communities do not carry adequate stocks of opioids.” (emphasis supplied). Of course, the study did not measure patient access at all but simply assessed availability. As well, the editorial authors appear to consider only Hispanics, and particularly blacks, as minorities being of concern; yet, it is those immigrants from Asia who have limited or no English, Spanish, or other common United States language skills and have the greatest potential for problems with access. Further, the editorial authors put part of the “blame” for limited access upon physicians in minority neighborhoods; yet, there was no indication in the article that the patients who need the opioid agents were seen exclusively by physicians in minority neighborhoods. Indeed, at least the study authors who see cancer patients for palliative treatment are affiliated with Mt. Sinai Hospital in New York; the neighborhood directly surrounding the hospital (5th Avenue and 101st Street) can arguably be considered to be outside the “minority neighborhoods”. As well, if such a contention were true, then the editorial authors’ primary bias—that “the common thread of the [study and other] findings is a subtle form of racial bias on the part of medical care providers”—would implicate racial discrimination by minorities against members of their own race, because physicians who practice in minority neighborhoods are often members of the same minority. Indeed, although the editorial authors state that “Physicians . . . must learn to see people not through the lens of race but instead as the individual persons they are,” it is the editorial authors themselves, who have mischaracterized the study they have been asked to comment upon and have immediately played the racial discrimination card in interpreting the study findings, who are guilty of first seeing patients and physicians as members of a particular race rather than looking for explanations with at least comparable plausibility. Although it is often difficult not to look through tinted lenses of one’s past if one has been subjected to any racial discrimination, 4 the editorial authors’ admonishment assumes that others, and not themselves, act in the manner that they so abhor.

Overall, the study provides interesting information on the availability of opioids in a variety of neighborhood pharmacies in New York and points toward potential concerns regarding access to appropriate pain relief. However, the accompanying editorial utilizes hyperbole when attempting to fit the findings into a racial discrimination causality claim.

Bryan A. Liang M.D. Ph.D. J.D.


1. Liang BA. Deselection under Harper v. Healthsource: a blow for maintaining patient-physician relationships in the era of managed care? Notre Dame Law Rev 1997; 72:799–861.
2. Liang BA. The practical utility of gag clause legislation. J Gen Int Med 1998; 13:419–21.
3. Posner RA. Economic Analysis of Law. Ed. 5. New York: Aspen, 1998.
4. Liang BA. Race cards: why justice and fairness can never be color blind. San Francisco Daily J 1995;Nov 2:6.
© 2001 Lippincott Williams & Wilkins, Inc.