FACTORS AFFECTING OLDER PEOPLE’S WILLINGNESS TO PARTICIPATE IN RESEARCH
To begin this section, several recruitment lessons were offered to increase the participation of older community-dwelling participants in frailty/injury prevention studies that appear relevant to the current research (Ory et al., 2002). These include:
In addition, Ory et al. (2002) state that recruitment yields can be bolstered by working through people living in residential apartments associated with a long-term care facility. Further, many older people temporarily relocate to warmer climates during the winter months, so that needs to be taken into account when designating recruiting windows. Finally, they state that it is more efficient to accept a refusal rather than invest the effort necessary to keep an uninterested participant in the study.
Wong (2004) reported on a polypharmacy intervention by Premera Blue Cross in conjunction with the Washington State Medical Association and the Washington State Department of Health that used the brown-bag method. The intervention included: (1) a brochure mailed to members that explained the potential risk for drug interactions when multiple medications are prescribed; (2) a medication log for members to list their medications by name, strength, directions for use, and purpose; (3) a brown bag in which members were directed to put all medications from all providers (prescription, nonprescription, and herbal) and to take to their provider; and (4) a letter mailed to providers explaining the Polypharmacy Program, and including the member brochure and the brown bag. The purpose of the medication review was to ensure that members were taking the correct medications at the appropriate doses, reduce medication duplications or interactions, and help to reduce potential adverse drug reactions. A public relations campaign was initiated to increase the effectiveness of the program by reinforcing the need for multiple-medication evaluations and to reach many individuals in the community. The PR campaign included television interviews on evening news programs, newspaper articles, and discussions on morning radio programs.
In 2001, polypharmacy material was sent to 12,000 members over age 55 who had at least five or more separate medication claims in one quarter for drugs on a custom list of medications defined as maintenance medications. In 2003, the Polypharmacy Program was expanded to include all members over age 18. As of February 2004, material had been mailed to 48,700 members and 5,000 providers in Washington, Alaska, and Oregon.
The intervention was evaluated through annual member and provider surveys. The member surveys were mailed four months after the polypharmacy mailing and consisted of questions addressing member recall of receipt of the program intervention material, and whether they read the material, completed the log, and completed a medication review with their physicians (and if so, what types of changes, if any, were made to their medication regimes). The polypharmacy material recall rate was 48 percent among members who responded to the survey. Of these, 88 percent indicated that they read the material, 36 percent completed the medication chart, and 50 percent had a medication review within four months of receiving the mailing. The study found that over two-thirds of patients (68 percent) received prescriptions from multiple physicians, and one-third of patients received prescriptions from three or more doctors. Overall, 52 percent of the respondents who recalled receiving the materials rated the program helpful or very helpful. Of those who had not visited their physician at the time of the member survey, 53 percent indicated that they plan to take their medications to their physician for review. Among those who brought their medications to their physician, one out of three had some type of medication change (add a medication, stop a medication, or change a dose). Of the providers who responded to the survey, 37 percent agreed that the program helps to reduce medication-related complications and hospitalizations. A pharmacy claims analysis indicated that in the year prior to the intervention, there was an average increase in the number of medications per person from 4.5 to 5.5. In the year following the intervention, the number of medications per person decreased from 5.5 at the time of the intervention to 5.0 after one year.
Farris et al. (2004) evaluated participants’ satisfaction with the reviews conducted during the Iowa Brown Bag program. Of the 24,825 Iowa Priority enrollees, only 3,198 (13%) sought and received a review in the year 2002. Enrollees were made aware of the review program through enrollment material, the program Web site, quarterly newsletters, and pharmacists. Enrollees who participated in a review completed a survey regarding motivation for obtaining a review; medication changes related to the review; satisfaction with the review; and willingness to pay for the review. The most common reason provided by the Farris et al. (2004) study respondents for participating in the review was to save money on prescriptions (65%), followed by obtaining more information (21%) and to ensure that all medications were safe to be taken together (20%). On a scale of 1 (very dissatisfied) to 7 (very satisfied), respondents had an average satisfaction rating for their review of 5.1, and 68 percent rated it as a 5 or higher. The percentage of respondents indicating willingness to pay some part of the cost for such a review was 48 percent for an initial review, and 29 percent for an annual review. When asked if they would have an annual review, 24 percent responded that they “definitely” or “probably” would. Survey respondents who did not seek a review provided the following reasons: they did not need a review (29%), they did not know about it (29%), or their physicians or pharmacists were not providing the reviews (11%).
Fillit et al. (1999) conducted a prospective study with a follow-up survey to examine the effects of a brown bag review by primary care physicians on prescriptions written for elderly members of a Medicare-managed care organization who were at risk of polypharmacy. Prescription drug data were obtained for 37,372 members of the Houston, Texas, area of the NYLCare 65/Gulf Coast Medicare risk health maintenance organization. Members identified at risk of polypharmacy—defined as those who had received 5 or more prescriptions during a three-month study period—were mailed a letter suggesting they make an appointment with their primary care physician for a medication review, because an electronic review of their medications indicated they were receiving multiple medications. A brown bag was provided with a label that read, “Request a medication review with your doctor. Bring your prescription and nonprescription medications to your doctor.” Each primary physician in the plan received a mailing containing an introductory letter, a member-specific medication report, and clinical practice guidelines for managing polypharmacy. The medication management report for each of their patients at risk for polypharmacy listed all prescription medications captured by the administrative claims data (only those prescribed for at least 30 days), as well as the dosages and number of pills dispensed per prescription. Use of over-the-counter medications was not available in the administrative database. A follow-up survey was sent to each member considered at risk for polypharmacy and to all primary care physicians treating one or more of the members at risk for polypharmacy. The member questionnaire was anonymous and contained 36 items: 3 concerned the receipt of the brown bag and resulting actions, 7 concerned the conduct of the medication review, 11 concerned symptoms and health care usage related to medication side effects, 1 item concerned changes in overall health, and the SF-12 instrument used to measure physical and mental health status.
Of the 37,372 member 65 or older enrolled in the Program, 5,737 (15%) were at risk for polypharmacy and were mailed the brown bag material, and later the member questionnaire. Of the 5,737 questionnaires mailed, 2,615 were returned. Of the 2,615 members who returned questionnaires, 1,087 (42%) had participated in the medication review with their physicians. Ninety-six percent of the members who participated in the medication review reported that they had a discussion concerning prescription medications and 72 percent indicated that nonprescription medications were also discussed. As a result of the medication review, 20 percent indicated that the doctor stopped or discontinued a medication, 29 percent indicated that the doctor changed the dosage of medication, 11 percent indicated that the doctor discovered medications purchased without a prescription that the doctor did not know the patient was taking, and 17 percent indicated that the doctor discovered medications prescribed by another physician that the first doctor did not know the patient was taking. Members who participated in the medication review were slightly more likely to have worse physical and mental health function scores on the SF-12 than those who did not request a medication review.
Other researchers have found the rate of self-referral for a brown bag medication review to be low. Nathan, Goodyer, Lovejoy, and Rashid (1999) distributed flyers to all households within the health authority localities of Bexley and Greenwich in southeast London, posted fliers in the 23 pharmacies within these health authorities, and trained pharmacists, nurses, and general practitioners to identify and recruit candidate subjects. Although the service was open to any member of the public, only 205 patients volunteered to participate. Ninety percent of the volunteers were actively recruited by the pharmacists. Once recruited, subjects were issued a brown bag and were asked to bring all their medication from home, including prescribed and over-the-counter medications, medications out-of-date or no longer used, and any reserve supplies. In this study, the resident pharmacist conducted the reviews, which took approximately 30 minutes each. Medications were reviewed to identify contraindications, interactions, adverse reactions, and problems with administration and compliance. The pharmacist explained the purpose of medications, and provided advice on dosage, administration, or compliance as necessary. With patients’ consent, pharmacists reported any medication-related problems to the patients’ general practitioner.
Follow-up interviews with patients indicated that, although they had initially been wary of the process and suspicious of the motives, they were unanimously in favor of medication reviews and thought that they should become a permanent service. Patients offered that they would have been more willing to participate had their doctor recommended that they do so. Several pharmacists recommended that closer collaboration with general practitioners was necessary to ensure the success of this type of activity. It was also recommended that pharmacists conduct the reviews in the general practitioners’ surgery, as it was a more suitable environment and there could be much better liaison with the doctor—this was highlighted by the need for more information by the review panel ranking medication problem severity, which was not available to the reviewing pharmacists. Additionally, pharmacists were reluctant to provide information to general practitioners out of concern that it would be perceived as an intrusion into the general practitioner’s area of clinical responsibility. The few general practitioners who responded to a follow-up survey indicated that they were in favor of the medication review activity.
Two focus groups were conducted to determine patients’ views of the pharmacist-led medication reviews conducted within Greater Glasgow Primary Care Trust, described above, which began in 1999 (MacRae et al., 2003). Between 1999 and 2003, over 20,000 patients participated in the review service. The 14 focus-group participants were sampled with respect to age, gender, and number of medication changes resulting from the medication review. Focus group members were asked to provide their views about their initial expectations, positive and negative aspects of the medical review process, and potential areas for future improvement. The positive aspects of the reviews included obtaining explanations about medications and knowing the medication regime was optimized as a result of the review. Problems arising as a result of the medication reviews included concerns about changing medications, the potential to upset the physician-patient relationship, and apprehension regarding physician rationale in the past for issuing or not issuing a medication under recommendation by the pharmacist as a result of the medication review. A majority of the focus group members stated that pharmacist-let medication review should be provided on an ongoing basis; however some patients wanted reassurance that pharmacists and physicians were working collaboratively.
Levenson (2002) reported on five focus groups attended by a total of 26 women and 15 men over age 60 in the United Kingsom as part of the Medicines Partnership Program to understand their views regarding the medication review standard published in 2001 by the National Service Framework for Older People. Key issues for discussion included: whether participants had heard of medication reviews; whether they had had a medication review; whether their physicians routinely ask about how they are doing with their medications; whether they always take their medications as directed; whether they had their medications reviewed within the past year, and if so how the review went, and who was involved; whether they were asked to take their medications with them; what they would like to discuss in an upcoming medication review; with whom and where they would like to have their medications reviewed; types of information they would like to have in advance of a review; whether they would like a written discussion about their medications; and what health professionals could do to help them get the most out of a medication review. Participants in each group suffered from a range of long-term medical conditions, and many reported regularly using more than four prescribed medications. Several also indicated that they took over-the-counter medications, vitamins, and homeopathic remedies. Although most people in the groups indicated that they comply with the advice given in taking medications, their comments and anecdotal stories indicated otherwise. Many indicated that they stopped taking medications or reduced dosages in response to unpleasant side effects. In one group, all participants indicated that they altered their medication routine from what was advised, because they felt overmedicated.
Of particular interest to the present research is what participants in the focus groups conducted by Levenson (2002) thought about where medication reviews should take place and who should be involved. In general, there were no strong views regarding where a review should take place, however, there was support for the opinion that medication reviews should be conducted face-to-face as patients would be flustered by telephone communication about their medications. Regarding who should be involved in medication reviews, general practitioners, hospital doctors, pharmacists, and to a lesser extent nurses were suggested as possibilities. The consensus of the groups was that the quality of the review and the attitude of the reviewer were more important than the professional background. Most important was that people wanted someone who was knowledgeable and had time to talk and listen. Those who had good relationships with their general practitioners felt strongly that the GP or possibly the practice nurse should conduct the review. Others had low expectations about GP review because their GPs had long waiting times for scheduling appointments and long waits at the office, and given those conditions, they were doubtful that GPs could add medication reviews to their commitments. Pharmacists were considered as knowledgeable and participants were open to the possibility of pharmacist involvement, provided that there was good communication of the results to the physician. Nurses were viewed as less informed about medications and therefore less suitable to conduct reviews.
Although most of the focus group participants had not participated in a medicine review, those who had participated found the experience to be helpful. The majority of those who had not experienced a medicine review indicated that they would appreciate the opportunity to discuss their medicines with a health professional. Those who stated that they would not want to participate in a review provided the following reasons: (1) avoiding unnecessary change (“I say, if there is nothing wrong, don’t mend it”), (2) fear of being taken off a medication that they depend on (“if they suggested coming off a pill—diazepam or hormone replacement treatment—it would be difficult to come off”); and (3) concern that the medication reviews may be to save the NHS money and not primarily for the benefit of patients.
Zermansky et al. (2001) conducted a study to determine whether a pharmacist can effectively review repeat prescriptions through consultations with elderly community-dwelling patients. Proposals in the United Kingdom have recommended that pharmacists review medication use in light of the fact that general practitioners’s workloads are already on the increase, and based on studies in the United States in the 1990s showing the benefits of pharmacists reviewing long-term prescriptions in community practice. Two earlier studies in the U.K. showed that pharmacists identified more drug-related problems than normal care (Granas and Bates, 1999; Mackie et al., 1999).
The Zermansky et al. study design was a randomized controlled trial of clinical medication review by a pharmacist against normal general practice review. Four general practices in Leeds Health Authority with four or more partners were recruited by random selection. The participating practices provided lists of registered patients 65 and older who received at least one drug on repeat prescription on June 1, 1999. Subjects were excluded if they were in nursing homes, had a terminal illness, or were in a clinical trial. Eligible patients were sent a letter requesting their participation; those who consented were randomly assigned to either a control group (normal care review) or an intervention group (clinical review by pharmacist). The intervention was conducted in the pharmacist’s clinic, although immobile patients were visited at home, and consisted of a discussion with the patient about conditions being treated and relevant symptoms; identification of drugs being taken; identification of indications for which they are being taken; assessment of adherence; an evaluation of continuing need for the drug, suboptimal treatments, side effects, drug interactions and contraindications, and cost considerations; and an evaluation of whether major, minor, or no change is needed. Patients in the control group continued to receive normal care from their practitioner and were called in for review of treatment by their general practitioner according to normal custom in the practice. Only 44 percent of the control group patients had a documented review with a doctor. The measures of effectiveness were the number of changes to repeat prescriptions during the 12-month study period, changes in the number and cost of medicines, changes in dose frequency, and effects on healthcare workload (general practitioner consultations, hospital outpatient admissions, and acute admissions).
Results were reported for 1,131 patients with a mean age of 74. The mean number of changes in at least one repeat prescription per patient was 2.2 for the intervention group and 1.9 in the control group. This difference was significant. Changes included starting a new drug, stopping a drug, switching drugs, changing doses, changing to a generic, changing formulation, or changing frequency. The only type of change that was significant was that more patients in the control group started taking a new drug. However, a higher percentage of treatment group patients were advised to stop a drug than control group patients (41% compared to 33%). The authors point out that stopping unnecessary medications is important for two reasons: patients’ compliance tends to decrease with increases in the number of drugs; and stopping unnecessary drugs reduces the risk of adverse effects and interactions. Although the number of drugs and costs rose in both groups over the 12-month period, the increase was significantly less in the intervention group. The intervention group showed no evidence of adverse health outcomes as measured by the need for consultation with a general practitioner or hospital treatment.
The authors mention that only half of the contacted patients were successfully recruited. In a follow-up study to determine why older patients declined to participate in the pharmacist-led medication reviews, Petty et al. (2001) found that consenting patients were dissimilar from consenting patients. Patients were less likely to participate in reviews if they were older or were female. The authors comment that the older patients may have found it more difficult to attend if they were “unwell” or were more likely than younger patients to have difficulty reading or understanding the letter. Females may have been less likely to participate because, as a rule, they visit their doctors more frequently than men and may have felt that they have ample opportunities to discuss medication use. Patients were more likely to consent if they were in the group taking 5 or more repeat medications (as compared to the group taking from 1 to 4 repeat medications). The researchers state that this may be due to hopes that the review would uncover medications that could be stopped, or because people taking fewer than five medications may see the review as a waste of time.
During follow-up telephone calls to nonconsenting patients, 10 themes emerged describing reasons for nonparticipation. First, several readability and comprehension difficulties were uncovered in the letter of invitation to the medication review. The font used for the letter was Times New Roman size 12. Patients with poor eyesight stated they had difficulty reading the letter, and no one in the home to read it to them. Others misinterpreted the location for the medication reviews, thinking they had to go to the researcher’s university or local hospital instead of to their own physicians’ practice, even though the letter was written on the physician practice letterhead. A subset believed the study involved a clinical trial, despite wording in the letter stating that the study was not a trial. Others stated that they took no medicines, due to a belief that inhalers and creams did not constitute medication. Second, a subset of invitees was unreachable either by telephone or mail. Third, cognitive impairment resulting in confusion about the study and the appointment explained some of the failure of invitees to participate. Fourth, some of the patients described themselves as “unwell” or just released from the hospital, and did not want to participate. Fifth, a subset was unavailable because they were either unwilling to change their routines, were in a hospital or daycare setting, or were moving out of the area. Sixth, several patients were uncomfortable with having a pharmacist review their medicines, thinking it could jeopardize the patient-physician relationship, while others explained they had just had their physician medication review. Others were suspicious of the motives and believed that the researchers were really checking up on the physician. Seventh, some patients were happy with their medication regime and did not want to be advised to make changes or stop medications. Eighth, patients with only one repeat medication thought it would be a waste of their time and the pharmacist’s time. Ninth, some patients have a general negative feeling about the health care system, and didn’t want to be bothered any more than necessary. And finally, some of the patients thought the purpose of the review was to save the government money by changing their medications to cheaper alternatives, and they felt mistrustful.
Petty, Knapp, Raynor, and O’house (2003) conducted focus groups to determine consenting patients’ views about pharmacist-led medication review clinics run in their general practitioner’s offices, and to identify barriers to participant participation. The mean age of the 18 participants was 73, with a range of 67 to 86. A range of socioeconomic backgrounds was represented, although none of the study practices represented rural communities. Ethnic minority groups were not recruited. Eleven of the 18 participants were males. The mean number of repeat medications they took was 5.5, ranging from 2 to 10. Perceptions before attending the clinic ranged from positive (an opportunity to gather information about the medications they were taking) to suspicion that the reviews were being conducted to save money. After the review, many found the review informative and reassuring, while others were disappointed with the outcome of the review (e.g., they hoped they would be advised to stop long-term medications, but were not so advised). There was also some feeling that there is a hierarchy of who is responsible for managing patients: a general practitioner should not change treatments prescribed by a specialist, and a pharmacist should not change treatments prescribed by a general practitioner.
Kriska and Ross (2002) found that when general practitioners in England received training in the systematic approach used by pharmacists for medication reviews, they were better able to identify pharmaceutical care issues (PCIs) than when they reviewed case notes alone. Following training by an experienced clinical pharmacist, general practitioners and community pharmacists performed medical reviews on patients age 75. Nurses received training in a different review process to help them identify medication-related problems during an annual health check. General practitioners documented 2.7 PCIs per patient, community pharmacists 1.6 per patient, and nurses 1.0. The pharmacists and nurses made significantly fewer changes to prescribed drug therapy and monitoring than the general practitioners. The authors note that this finding may have been the result of higher general practitioner confidence in effecting changes and requesting monitoring. Follow-up questionnaire responses from the patients indicated that the reviews were useful (77%), and 42 percent identified a preference for a general practitioner review or continued general practitioner management. Kriska and Ross reported that the participation rate in their study was higher than that found in controlled studies of pharmacological review. They propose that this reflects a preference expressed by patients for review by general practitioners.
Jainkittivong, Aneksuk, and Langlais (2004) reported on the medical conditions and medications used in a sample of 510 Thai dental patients 60 and older. As dentists will be treating more elderly people in the future, and older people have more medical conditions and take more medicines than younger people, the authors highlight the importance of obtaining a patient’s total health history before undertaking dental procedures. This is because some systematic diseases may influence oral health and/or dental treatment to some degree and dental treatment may have an influence on some systematic conditions. The authors include a list of medications with possible adverse effects of dental significance. In the Jainkittivong et al. (2004) study, one investigator interviewed subjects regarding their medical histories and medications. The subjects’ family members were interviewed in cases where the subject’s recall was poor. In addition, medications were verified whenever possible by inspection of medication container labels. In light of their findings that 68.3 percent of their subjects 65 and older took medications, and the average number of drugs increased with increasing age, the authors state that taking of a thorough medical history is especially important among older people, and should include a health questionnaire, an interview, copies of treatment records from physicians, and a careful analysis of all drugs being taken. Medical consultations with patients’ physicians may also be needed for patients who provide conflicting or vague medical histories. The results of this study point to dentists as a group of professionals who routinely collect medication use from their patients, and may be a resource for future NHTSA studies of medication use by older drivers.
In the study by Linjakumpu et al. (2002), information about medication use was collected in personal interviews conducted by a trained nurse in a health center. This study involved 1,131 community-dwelling people 64 and older in 1990/91 and 1,197 community-dwelling people 64 and older in 1998/99 in Lieto, Finland. In the first survey, only 7 percent of the invitees declined, did not respond, had moved elsewhere, or were unable to participate due to illness or disabilities. In the second survey 18 percent of the invitees did not participate for the same reasons. Subjects were asked to take their prescription forms and drugs with them to present to the nurse what they were currently taking. A close relative or caregiver provided information if the person could not answer questions, had dementia, or was not in good condition. If a subject could not visit a health center, a nurse made a home visit to check the medication in use. The brand names of all prescription drugs taken by the interviewee during the seven days prior to the interview was recorded. Both regular and irregular prescription drug use were recorded. Medication was defined as regular if it was taken daily, or at regular intervals, for example once per week or month. Irregular medication was defined as medication taken when needed. In addition to medication use, a questionnaire was administered to collect information about social background and quality of life. This study points to a relatively high participation rate (93% in the first survey and 82% in the second survey) of older community-dwelling people in medication use studies (albeit in Finland, as opposed to the United States) where nurses are the professionals recording usage.
A total of 7,543 noninstitutionalized elderly in the city of Randers, Denmark, were contacted by mail, and asked to complete a questionnaire on health problems, falls, exercise habits, and use of calcium and vitamin D, and were advised that a home visit would be offered within six months if they participated. This was a free-of-charge medication review by a nurse, a home safety inspection, an evaluation of possible health and food problems, and for some participants, distribution of free calcium and vitamin D supplements. Sixty percent of those contacted returned the completed questionnaire. Nonrespondents were recontacted twice by phone or by mail. Eighty-five percent of those who returned the questionnaire agreed to the in-home evaluation, for an overall acceptance rate of 51 percent. Females had a higher acceptance than males. Participation did not change significantly from ages 66 to 84, however, after 85, acceptance decreased significantly. Widowed people had the highest participation, starting at 62 percent among the youngest and decreasing to approximately 33 percent among the oldest. The trend was opposite for the never married, starting with 34 percent participation for the youngest and increasing to 47 percent at ages 95 to 99. Acceptance was highest (56 percent) when calcium and vitamin D were offered in combination with an evaluation of prescribed medicine (without the home safety inspection and health and food problems evaluation). Previous contact with the public service social center doing the recruiting was associated with higher acceptance than no previous contact, and more effect on acceptance than the type of intervention offered. The authors hypothesize that the low acceptance rate may have been influenced by the requirement that participants be willing to accept a visit in their own homes, which may have been regarded as unwanted.