Objective: To describe and evaluate the varying medication management practices and policies in Massachusetts assisted living settings. In particular, the project studied residents' quality of life with respect to medication assistance, appropriateness of medication regimens, and error rates of medication administration. These were analyzed as a function of variations in medication management practices. The goal was to better understand medication management issues in assisted living settings generally and to generate questions for further in-depth study. Design: Data collection was undertaken at all sites in Massachusetts which, at the time of the study, identified themselves as assisted living settings and were willing to participate. The research team developed a survey tool to gather information about their medication management practices and other operational practices in the participating sites. Medication administration was observed and rated for errors in a random sample of 102 residents, with use of an adaptation of the methodology used by surveyors in nursing facilities. Finally, an in-depth record review was completed on 23 randomly selected residents. This sample of residents also completed a resident satisfaction tool, designed by the research team to capture their subjective experiences and impressions regarding the quality of assistance they receive with their medications. Setting: Assisted living settings offer a range of assistance with activities of daily living in a more residential setting than nursing facilities. The twelve sites included in this study ranged in size from 18-283 beds. Participants: All residents in the participating sites were considered for selection in the study, and 144 residents were randomly selected; 102 residents gave informed consent to participate in the study. Of these, seven were cognitively impaired, and family consent was obtained. The reasons for non-participation included cognitive impairment and inability to obtain family consent (35) and absence from the unit during the study (7). Outcome Measures: Errors in right person, right drug, right dose, right time, with/without food, and technique were calculated, adapting the Health Care Financing Agency methodology. Resident satisfaction with medication assistance was surveyed. Results: An overall error rate of 3.62% was found. Non-professional direct care staff provided medication assistance with the same or lower error rates as staff nurses or outside agencies. Staff training and an ongoing relationship with a consulting pharmacist were correlated with the lowest medication error rates. Conclusions: The findings indicate that even in the unregulated industry of assisted living, facilities in Massachusetts were able to offer residents a high quality of service with a medication error rate lower than that found in comparable studies in nursing facility settings. Key Words: Assisted Living, Medication Error Rates, Medication Management. Abbreviations: HCFA = Health Care Financing Administration; OTC = over the counter. Consult Pharm 1998;9:1001-14.

This descriptive study captures medication management practices at a particularly interesting time and place in the history of assisted living as a long-term care setting, a time that provides insights for both regulators and providers as the industry matures. At the time that the data were collected, assisted living existed as an unregulated entity in Massachusetts. The handful of providers were experimenting with various aspects of their product from building design to service delivery.

Nationally, assisted living is a new field with diverse organization, design, and practice.¹ Underlying the medication management debate are two more fundamental debates:^2-4

Should assisted living facilities provide the full range of services, including nursing care that older people need to "age in place?"

Given that residents of assisted living facilities are often physically or cognitively impaired, how much responsibility should they be allowed to assume for their own well-being and how much oversight should the facility be providing?

Many providers and consumers believe that assisted living can offer the full range of services available in nursing facilities, but in a residential setting allowing for more resident autonomy but also more risk.⁵ Others believe that assisted living is most suitable for those who can direct their own care and that a limited range of services is appropriate in this setting.⁶

Because many residents who have no other nursing needs do need help with their medication, this area of assistance is often at the forefront of the debate on the limits of assisted living.⁷ We expected, and found, considerable diversity in philosophy and practice among the assisted living facilities surveyed in the unregulated environment that existed in Massachusetts at the time data were collected. A full description of this diversity can be found in the section on Findings Regarding the Surveyed Assisted Living Settings.

In general terms, the practices ranged from extreme informality to a complete medication administration policy. In the most "informal" sites, the staff did not track the names of residents’ medications and provided simple reminders at those times the resident or family had told them a reminder would be needed. At the other end of the spectrum, the medication management system was essentially the same as that found in most nursing facilities.

This report describes that diverse experience and provides data that can be used to evaluate the success of the various approaches assisted living providers were utilizing at the time of the study. The investigators hope this information will prove helpful to providers, consumers, and regulators working to foster resident well-being as regulations are written or revised and as the industry matures.

Study Sample
At the time the study was initiated, in the fall of 1993, 15 assisted living facilities were operating in Massachusetts. Twelve sites agreed to participate in the study. Of those that did not participate, one stated that their policy was not to participate in research and two were in transition with respect to key staff and medication management procedures at the time of the study.

All residents living in the participating assisted living facilities were eligible. A complete resident list was obtained from each facility, and individuals were assigned identification numbers. With use of a random number generator, a sample of 12 residents per site was determined. The smallest facility had only 12 residents; therefore, we sampled 12 residents from each site to be consistent. The sample size was determined to be adequate for this descriptive and exploratory study. One hundred forty-four residents were originally sampled. Forty-two of these residents were cognitively impaired, 35 of whom were therefore disqualified from the study; however, family consent was obtained for the remaining seven cognitively impaired residents. In addition, seven other residents were absent at the time of the study, making a total of 102 residents ultimately included in the study. Table 1 shows characteristics of the study population.

Data Collection
All data were collected between September 1 and December 15, 1993, by the research assistant, a registered nurse with over 20 years of clinical experience who was also a graduate student with training and experience in survey procedures. She made at least two visits to each residence. At the first visit, information was obtained on policies and procedures with respect to medication management, consent forms were distributed, and a meeting describing the purpose of the study was held with residents.

On the second visit, the research assistant reviewed the medication records and observed one, usually the morning, medication pass and then interviewed each

participating resident. More than half the medications were taken in the morning, and all were taken within a limited period. Only 12 residents took medications throughout the remainder of the day. We had no reason to suspect that error rates would differ at other times. While we had record of 405 medications that residents were taking, we observed only 389 medications actually being taken. Medication errors were observed and counted using a methodology modeled after that developed by the Health Care Financing Administration (HCFA) for nursing facility surveyors.⁸ Medications taken both with and without staff assistance were observed and all errors were counted.

Study Findings
Findings Regarding the Surveyed Assisted Living Facilities. The facilities exhibited a range of characteristics, which are found in Table 2. Additionally, one site was a low- and moderate-income senior housing project that had added assisted living services under the new state Group Adult Foster Care program.

The study found a wide range of procedures for residents in managing their medications. Some facilities characterized themselves as "residential hotels" providing meals and housekeeping, with other services provided either by the management or outside agencies on an as-needed, "unbundled" basis. Several of the larger facilities grouped residents who required additional assistance in activities of daily living or medication management and arranged for separate payment for these services. In these cases, only the assisted living portion of the complex was considered in our study. Other facilities identified themselves solely as assisted living, providing such services to most residents as part of the basic package and marketing to and admitting only residents who require some assistance with activities of daily living.

Medication services were categorized as follows: an outside agency handles medications for any residents who require such assistance; the assisted living staff provides medication reminders; the assisted living staff administers medications while characterizing their activity as reminders; the assisted living staff administers medications directly to residents. In some cases, residents in the last three categories who had other nursing needs requiring the use of outside nursing personnel used the same agency to assist with medications. In those facilities providing reminders or stating that they provide reminders, either family or outside agencies were often involved with setting up medications in daily or weekly boxes. "Medication administration" was defined as an assisted living staff member removing medication from its original pharmacy container out of the sight of the resident and placing the medications in daily or weekly boxes or souffle cups.

Written policies and procedures regarding medication management ranged in participating sites from no written policies to relatively complete written policies. The medication record systems also varied, with some systems being similar to those found in nursing facilities, and others being much more informal. The three sites that stated they provided medication administration kept full records of the medications for the residents involved, while among the other sites the range of detail in record keeping varied. Generally, assisted living facilities do not maintain detailed, up-to-date medication records for residents for whom they do not have medication responsibility.

The type of storage and the degree of locking and other security for medications also varied considerably, both among and within each site. Thus while medications were most often kept in ordinary pharmacy vials, some sites encouraged the use of weekly pillboxes, sometimes filled by families. Some residents used unit-dose bubble packs or compliance packaging. In many sites, we found a combination of these forms of medication storage, depending on the needs of the individual resident.

Finally, the facilities varied in the amount of involvement they had from a pharmacist. The level of involvement was determined by data collected from the facility questionnaire and policies and procedures, which included items listed in Table 3. Facilities that answered 2 of the following 3 questions (2, 3, and 5h.) "yes," were considered "high". Facilities that answered question 1 "no" and checked any of 5a.-g., were considered "low". Facilities that answered question 1 "yes," but otherwise checked 5a.-g., were considered "medium".

TABLE 3. Questions Regarding Level of Involvement of Consultant Pharmacist* Does your facility try to have most medications come from the same pharmacy? ____________ Does your facility require a drug regimen review for residents? ____________ If Yes, who performs the drug regimen review? ____________________________________ Who answers drug information questions that residents or staff have? _____________________________________________________________ Has your facility ever utilized the services of a consultant pharmacist? a. ______ No, not familiar with their services b. ______ No, familiar but don't use services because: c. ______ Can't afford d. ______ Not worthwhile e. ______ Not needed f. ______ Other g. ______ Yes, in the past, but stopped because: _______________________________ h. ______ Yes, currently used ______ visits per year. ______ Hours per visit. * Excerpted from the facility questionnaire used in the study.

Resident Satisfaction with Medication Assistance. Overall, residents were pleased with the amount of assistance they received, with 94% reporting that the help they received was "the right amount." Four percent felt that staff were overly involved in their medication assistance, and 1% felt they were not receiving enough help. When asked if they had ever experienced medication errors, 2%-3% of the residents reported errors in each of the areas of type, dose, or time of medication at some time during their stay. Nine percent reported they had missed a dose at least once because staff had not reminded them.

While most stated that they had adequate information about the medications they were taking, 80% said that they did not have information about potential drug interactions, and 76% did not have information on their medications' potential side effects. Eighteen percent reported having experienced side effects from their medications since moving to their current assisted living facility. Ninety-eight percent said they would talk to the assisted living nurse if they felt their medication might be causing a problem, while 81% would talk to a family member and 34% would contact their doctor. When asked how they obtained their medications, 32% said they purchased them themselves, 24% said the assisted living staff helped them, 21% said the outside nursing agency arranged for the medications, and 20% said family bought the medications for them.

Findings Regarding Medication Use and Error Rate. The participants were taking a total of 559 medications at the time of the study. A medication was considered as "being taken" if there was a written record that it was being taken on a regular or as-needed basis or if we observed it being taken during the medication pass. Table 4 describes the residents’ use of medications. Of the 54 residents who take their own medications, 9 reported taking medications differently than as prescribed and 15 reported that assisted living staff were not aware of medications they take.

During medication observations, the researcher was able to observe 389 medications being taken by residents, with or without staff reminders or administration. In calculating the medication error rate, we divided actual errors by potential errors. This is a widely accepted methodology in the field. The errors recorded were: Wrong Person, Wrong Drug, Wrong Dose, Wrong Time, With/Without Food, and Improper Technique (including route). For example, if a medication is supposed to be taken with food and was not taken with food, this counted as an error, while if there is no requirement that the medication be taken with food and the resident took the medication with food, this did not count either as an error or as correct. With this method, there were 2,070 opportunities for errors and 75 errors were made (Table 5), for an overall error rate of 3.62%.

The calculation of the error rate contains two potential limitations: not all sites maintained complete records of medications prescribed, and errors of omission, which we were unable to count, may have occurred. In addition, we do not know how the inclusion of a more representative sample of cognitively impaired residents may have influenced the error rate calculation.

Using a similar methodology, HCFA, which oversees and surveys nursing facilities throughout the United States, considers a medication error rate under 5% "acceptable." ⁹ A 1987 study looking at medication error rates in nursing facilities found a reported rate in a retrospective review of records to be 0.2% and on observation to be 9.6%.¹⁰ The few other studies available used such different methodologies and counted such different types of errors that comparison is not possible. There have been no comparable studies of medication error rates among community-living elders.

A major focus of this study was to understand whether the way in which medication assistance is given affects the medication error rate. The primary areas we examined were medication policy (i.e., does the facility use an outside agency, have an in-house reminder system, or actually administer medications?), use of an outside pharmacist consultant, and qualifications and training of those providing medication assistance.

The error rate was highest in the three sites that claimed to provide only medication reminders but were actually administering medications (see Figure 1). Error rates were lowest in the one facility that actually provided reminders only, with those providing in-house administration and those using an outside agency in the middle of the range of errors made. A rank-order analysis was also done to better understand this model. This analysis was determined by investigators to be sufficient, as this was a hypothesis-generating pilot study. While the findings of both analyses were non-significant because of the small sample size, the direction of the findings suggests that the analysis of true in-house administration was the most accurate.

The highest training level of staff directly involved with medication management was related to medication errors (see Figure 2). The lowest error rate (1.6%) was found in the two sites that used only resident assistants to provide medication assistance, while among the three sites that had a registered nurse on staff who actively assisted with medications, the error rate was 6.55%. We hypothesize that having the same staff members remind residents to take their medications and help them with other activities of daily living improves communication and reduces the likelihood of errors.

Pharmacist participation also seemed to be inversely related to the error rate, with those three sites that had the highest pharmacist participation in medication services showing a low error rate (0.38%), and those with no or minimal pharmacist involvement having the highest error rate (4.28%) (Figure 3).

Staff training was the area in which we saw the most marked difference (see Figure 4). Assisted living sites where staff received formal training in medication management had only a 0.24% error rate, compared with a 7.3% error rate in those sites where there was no record of a formal medication management training program. A site was scored as providing staff with formal training if they reported such training on completion of the site survey and if they could produce a copy of a training syllabus or training materials. Training in medication reminders, when it was provided, averaged only one hour.

Figure 1. Error Rate By How Medication was Handled* *n=number of opprotunities for error.

Figure 2. Error Rate by Highest Level of In-House Staff Involved with Medication* *n=number of opportunities for error.

Types and Appropriateness of Medication Taken by Residents

The most frequent types of medications administered were categorized as central nervous system (156 prescriptions), cardiovascular (123 prescriptions), and gastrointestinal drugs (83 prescriptions). Of the central nervous system drugs being used, 49 were psychotropics. The rest were primarily pain medication, including 76 cases of residents taking non-narcotic analgesics and non-steroidal anti-inflammatory agents.

The use of psychotropic medications has been a concern in the elderly and institutionalized populations. In this sample, 38% of residents were taking at least one psychotropic medication. This is strikingly similar to the psychotropic drug use in another residential care drug use study, which found that approximately 35% of the older residents used at least one psychoactive agent, and 30% of those residents received two to four different psychotropic medications.¹¹ It is important to note that this sample contains an uncharacteristically small proportion of people with dementia. Had they been included, the percentage taking psychotropic medications would probably have been considerably higher.

Of those taking psychotropic medication, one resident was taking four, one was taking three, five were taking two, and the remaining 32 were taking one psychotropic. In addition to the polypharmacologic nature of such use, the fact that 25% of all residents in the sample are taking anti-anxiety or sleeping medications may be cause for concern. The findings regarding the classes of psychotropic drugs are shown in Table 6.

Another area of potential concern to both residents and assisted living facility administrators is the use of over-the-counter (OTC) medication. The total number of OTC medications taken by this group of residents on a regular basis was 180 (see Table 7).

Figure 3. Error Rate by Pharmacist Participation in Medication Administration* *n=number of opportunities for error.

Sixty-four residents (62.7%) were taking cardiac drugs, with a total of 123 prescriptions, for an average of 1.92 (range, 1-7) cardiac drugs for those residents taking cardiac drugs. A breakdown of these medications is shown in Table 8.

To evaluate the appropriateness of the medication regimens, we selected the most complete and comprehensive medication records to review. Twenty-three residents were included in the sub-sample. Review of these drug regimens is consistent with what might be expected in the general aging population: The pathologic conditions suggested by the medication record included cardiovascular disease, anxiety disorders, depression, Parkinson's disease, gastrointestinal disturbances, hypertension, diabetes, ophthalmic disease, and genitourinary disturbances. The following potential problems were identified:

• OTC medications in this sub-sample accounted for almost one-third of the medications used. However, there was no information regarding the appropriateness of these drugs or whether they were taken with a physician's or the facility's knowledge.
• One resident was found to be taking contraindicated drugs.
• Several residents were taking combinations of drugs that require proper laboratory monitoring, which should have been included in their service plan. It was unclear whether there was appropriate monitoring of these residents and whether information such as laboratory test results were reported to the facility, physician, or individual resident.
• Potential interactions between OTC medications such as cimetidine and prescription drugs such as warfarin were also found.
• Some medications such as ranitidine and digoxin may be combined but must be separated by at least one hour. Without Medication Administration Records, there is no way to ensure the proper timing in dispensing of these drugs.

The 23 residents in the sub-sample had a number of risk factors for medication problems, such as multiple medications, multiple disease states, cognitive and sensory deficits, multiple physicians and pharmacies, and unsecured storage of medications. A lack of information systems to adequately assess appropriate use of medications is problematic in assisted living facilities.

Figure 4. Error Rate by Formal Staff Training Prior to Beginning Medication Responsibilities* *Facilities using outside agencies were excluded. n=number of opportunities for error.

Regulatory, Legal and Policy Issues
At least three sets of laws and regulations in Massachusetts govern the medication management services that residents of assisted living facilities may receive. The first is the law regarding provision of housing with services to elderly people in Massachusetts. At the time of the study, Massachusetts laws stated that any provider of housing with services to the elderly must be licensed as a Level IV Facility, also known as a "rest home." An advisory letter issued by the Department of Public Health in August of 1992 reinterpreted the law such that entities that provide housing and services do not need to obtain a license as long as the tenants had their own apartments, as defined by having a lockable door, ability to have visitors at will, and other criteria, and that the elderly tenants could choose from whom to purchase services. Thus the "landlord" could also be a service provider, in competition with families and outside agencies from which the resident could purchase services. To the extent that the building owner provides personal care services, such as help with bathing and dressing, there is no requirement for further licensure. This would include the simple reminding of residents to take their medications. Since that time, assisted living certification and regulation by the Executive Office of Elder Affairs has been instituted.

Involvement with medication administration triggers the terms of the Nurse Practice Act. With respect to medication, this act states that both registered and licensed practical nurses may "plan and implement nursing intervention[s] which includes all appropriate elements of nursing care, [including] prescribed medical or other therapeutic regimens . . ."¹² The act specifically states that the nurse may delegate nursing duties to unlicensed personnel with certain exceptions, which include medication administration. Thus the assisted living facilities that were using unlicensed personnel to provide medication administration were not in compliance with Massachusetts law. In the absence of assisted living regulation at that time and given the fact that unlicensed personnel were permitted to administer medications in group homes and similar settings, there had been no cases of unlicensed assisted living facilities being charged with irregularities regarding their medication practices.

Given the ambiguity of the situation, some assisted living facilities interpreted the law to allow full medication administration by their own staff, so long as the staff member is a nurse. Others believed the law allowed for medication administration delegated to unlicensed staff. Yet other assisted living entities did not wish to or did not believe they were legally permitted to provide such services themselves, and used an outside agency for medication services. Finally, others restricted themselves to reminders.

The definition of "medication reminder" is also problematic. Clearly, if residents have full control of their own medications and the staff simply says, "It is time to take your medicine," there is no question that that constitutes a reminder. The one assisted living facility in the study that adhered most closely to the policy of giving reminders only in fact had a very stringent record-keeping system and stayed in the room with the resident to observe and record the taking of all reminded medications. That facility had no medication errors at the time of the study observation, suggesting that such a system can in fact serve residents well. More commonly, sites do one or more of the following medication-related activities: store medications, assist with opening pharmacy containers, provide water or juice with which to take the medications, check that the correct dose of the correct medication was taken, and keep records of the medication taken.^13,14 Just as clearly, if the staff touches the medication, takes it out of the pharmacy container out of the sight of the resident (such as to put it in a medication cup), hides the medication in food, or otherwise takes control of the medication, that would constitute administration. Providing the resident as-needed medications at the discretion of staff would also constitute administration.

However, at the time of the study there was ambiguity regarding what would be permissible. For example, at that time, the state agency preparing to regulate assisted living believed that the use of unit-dose cards would constitute administration. These are individually packaged pills on bubble-pack cards prepared by the pharmacist with a foil backing that can be easily punched out. This almost foolproof system makes it easy to know which pill to take at which time on which day and to verify after the fact that a medication has been taken. Other studies have compared the use of unit-dose to traditional medication vials and found the unit-dose methods to be associated with lowered error rates.¹⁵

A third set of regulations that currently govern medication management in Massachusetts pertains to the state Medicaid program, known as Group Adult Foster Care. This program requires nursing supervision for personal care services it reimburses. The regulations association with the program may influence the ability of assisted living settings to use unit-dose cards.

Another common practice is the use of daily or weekly medication boxes. These plastic boxes, available at all drug stores, allow consumers to take their medications out of the original containers and put all the medications for a particular day or time together. Like the unit-dose bubble-packs, these make it easy to know which medications to take at what time and to verify whether the medication has been taken. Consumers or a family member may place their medications in the boxes. However, the placing of medications in boxes certainly constitutes medication administration when done by a non-family member and, in Massachusetts and some other states, is defined as dispensing, an activity reserved to pharmacists. Although we do not have sufficient data to substantiate this, our impression is that when elders are taking multiple drugs, especially if they are visually or cognitively impaired and cannot themselves keep track of which medications they are taking, the use of such boxes may increase the incidence of medication error. From a regulatory point of view, the filling of boxes by unlicensed staff raises issues regarding the regulations governing the provision of pharmacy services. As this manuscript goes to press, the Massachusetts Board of Registration and Pharmacy is reviewing the definition of "dispensing" with respect to these boxes, as well as the issue of nurse delegation.

Conclusions
The findings of this study confirm that assistance with medications is an important service that is highly valued by residents of Massachusetts assisted living facilities. Overall, these facilities are performing this service with a lower medication error rate than the rate that would be found acceptable in nursing facilities and lower than that found in the few comparable studies done in licensed settings. Assisted living facilities that use non-professional direct-care staff rather than outside agencies or even their own staff nurses appear to be providing at least as good service in this area. Training of staff regarding their medication management duties was correlated to the lowest medication error rates and may be the most promising area for regulation.

Residential care regulations in many states prohibit assisted living and similar settings from providing nursing services, but they have excluded medication administration from this prohibition. By allowing true administration and requiring professionally recognized systems of training, record keeping, storage, and other policies regarding medication management, the state could ensure the appropriateness of this service.¹⁶ Nationally, regulations vary tremendously. States such as Massachusetts only allow assisted living facilities to cue or remind residents of medications, while in New Jersey, non-nursing staff may not only administer medications, but may also inject insulin.

Unlike nursing facilities where facility-related physicians monitor all patient care, assisted living residents typically retain their own physicians and may have one or more specialists who may or may not coordinate care among themselves.¹⁷ Given the risk factors for medication-related problems in this population, such as multiple drugs, multiple diagnoses, use of OTC medications, and medications uncoordinated by multiple medical specialists, there may be an important role for assisted living facilities to coordinate residents’ medication regimens beyond giving medication reminders or administering drugs.^18,19 This coordination may be undertaken by a well-trained staff nurse, consulting physician, consulting pharmacist, or team that uses all these specialties.

It should be noted that those sites that had an ongoing relationship with a consulting pharmacist had the lowest medication error rates. In addition, as part of the assisted living team, these professionals may provide assessment and coordination of medication regimens. In order to provide this service, consultant pharmacists must have training in clinical pharmacy and pathophysiology and a thorough understanding of the changes that occur with aging as related to medications.

The impetus for funding this research came in part from the release of the 1992 Congressional report, Drug Abuse and Misuse in America's Board and Care Homes: Failure in Public Policy, which found a lack of accepted procedures with respect to the medication assistance that many assisted living facilities provide to their residents.^20-22 Our study found that even in the absence of regulations, most assisted living facilities in Massachusetts were able to offer their residents a high-quality service. As Massachusetts and other states write or rewrite regulations for the evolving residential care industry, regulators and providers will need to find ways to provide reliable medication assistance to those residents who need such help, while maintaining the relative affordability of such settings and their emphasis on residential lifestyle and resident autonomy.

References
1. American Health Care Association. The long term care survey. Washington, D.C.: Health Care Financing Administration; 1991.
2. Baggett SA. Residential care for the elderly. New York: Greenwood Press; 1989.
3. Wilner MA, Bumpus S. Resident choices in assisted living: A case study. Presented at the Annual Meeting of the Gerontological Society of America, 1989.
4. Moon M, Gaberlavage G, Newman SJ. Preserving independence, supporting needs: The role of board and care homes. Washington, D.C.: Public Policy Institute, AARP; 1989.
5. Carley M. Unnecessary drug requirements. Contemporary Long Term Care 1992;15(12): 68-70.
6. Echevarria K, Skinner J. Clinical factors associated with facility rating of quality of life in board and care homes. University of South Florida (unpublished manuscript).
7. Wilner MA. Quality in nursing homes and residential facilities. Report to the AARP; 1992.
8. Health Care Financing Administration. State operations manual, provider certification. Interpretive Guidelines Section 483.25 (m) Medication Errors. Washington, D.C.: Department of Health and Human Services (HCFA Transmittal Number 274); 1995.
9. Health Care Financing Administration. Interpretive Guidelines for Medication Errors: State operations manual, provider certification. Washington, D.C.: Department of Health and Human Services (HCFA Transmittal Number 250); 1992.
10. Shannon RC, De Muth JE. Comparison of medication-error detection methods in the long term care facility. Consult Pharm 1987 Vol. 2;148-51.
11. Larrat PE, Spore D, Mor V, Hiris J, Hawes C. Medication utilization in board and care facilities. Consult Pharm 1995;10:1267-77.
12. Massachusetts General Laws, Chapter 112, Sections 74; 244 CMR 3.00, and following.
13. Mollica RL, Ladd RC, Dietsche S, Wilson Kb et al. Building assisted living into public long-term care policy: A guide for states. Portland, ME: Center for Vulnerable Populations: The National Academy for State Health Policy and Brandeis University; 1992.
14. Garrard J, Cooper SL, Goertz C. Drug use management in board and care facilities. Gerontologist 1997;37: 748-56.
15. Barker KN, Mikeal RL, Pearson RE, et al. Medication errors in nursing homes and small hospitals. Am J Hosp Pharm 1982;39: 987-91.
16. Regnier V, Hamilton J, Tatabe S. Best practices in assisted living: Innovations in design management and financing. Los Angeles, CA: National Eldercare Institute on Housing and Supportive Services, Andrus Gerontology Center; 1991.
17. Hyde J. Regulations and interpretation of regulations that interfere with the design and operation of special care units. In Special Care Units for People with Alzheimer's and other Dementias. Washington, D.C.: Office of Technology Assessment, U.S. Congress; 1992.
18. Avorn J, Dreyer P, Connelly K et al. Use of psychoactive medication and the quality of care in rest homes: Findings and implications of a statewide study. N Engl J Med 1989;320:227-32.
19. Wilson KB. Developing a viable model of assisted living. In P Katz, RL Kane, M Mazey (eds). Advances in Long-Term Care. New York:: Springer Publishing Co; 1993.
20. U.S. General Accounting Office. Drug use and misuse in America's board and care homes: Failure in public policy. Gaithersburg, MD: U.S. General Accounting Office; 1992.
21. U.S. General Accounting Office. Board and care homes: Elderly at risk from mishandled medications. Gaithersburg, MD: U.S. General Accounting Office; 1992.
20. U.S. General Accounting Office. Board and care: Insufficient assurances that residents' needs are identified and met. Gaithersburg, MD: U.S. General Accounting Office; 1989.