Question: Does medicine review increase shared decision-making or adherence?

Grading: 1+Well-conducted meta-analyses, systematic reviews of RCTs, or RCTs with a low risk of bias
Begley S;Livingstone C;Hodges N;Williamson V;
Impact of domiciliary pharmacy visits on medication management in an elderly population
Ref ID 75551997
Study TypeRandomised Controlled TrialFundingNot reported.
Number of participantIntervention group n=61; control group (V) n=63; and control group (NV) n=66.
Inclusion/Exclusion CriteriaInclusions criteria: to be aged 75 years or older; prescribed three or more different drugs; at least a twice daily dosage for one or more of the drugs; under the care of a participating consultant; consented to participate in the study; and was returning to their home (not further institutional care).
Patient CharacteristicsMajority of the patients were female (61% in the intervention group; 65% in the V group and 56% in the NV group). The median ages were 84 years (range 75 to 94) for the intervention group, 81 years (range 75 to 96) for the V group, and 82 years (range 76 to 92) for the NV group.
RecruitmentThrough discharge prescriptions were presented in the hospital pharmacy (provided they met the inclusion criteria). These were three hospitals from the Crawley and Worthing district health authorities.
SettingHospital pharmacies.
Interventions/Test/Factor being investigatedGroup A receiving home visits and counselling, group B which was the control and received visits only (called V group), and group C was the control group that received traditional pharmaceutical services with no visits except for the beginning and the end of the study (NV group).
Structured patient interviews were conducted during the domiciliary visits and consisted of six sections: patient information; drug knowledge; patient dexterity; abbreviated mental test; medication management; and compliance with medication regimen. Patients were seen during 12 months.
Other strategies were employed for improving patient compliance: emphasising the importance of compliance; giving clear instruction on the exact treatment regimen, in writing if necessary; arranging dosing times to fit into the patients daily routine; recognising the patients effort to comply at each visit; and simplification of the regimen if necessary.
ComparisonsBetween treatments.
Length of Study/Follow-upUp to 12 months.
Outcome measures studiedNo. of drugs prescribed and purchased; drug knowledge scores; patient dexterity scores; abbreviated mental test scores; medication management; compliance with medication regimen; contact with gp and health workers.
ResultsAt each visit there were significant differences between the groups in terms of distribution of patients at the various levels of compliance (p<0.001).Compliance was higher at 3 months and 12 months for the intervention group compared to the other control groups (p<0.001), despite the low compliance value for the intervention group at the 12 month visit.
Patients in the intervention group who increased their compliance rates between visits also increased their drug knowledge scores (p<0.005).
Mean scores for drug knowledge did not differ significantly between the groups at any of the visits, although the mean score for the intervention group increased significantly between the initial and the two weeks visits (p=0.001). There were no changes for patient dexterity scores between groups at any point of the study.
The intervention group did not report any significant changes in abbreviated mental test score, but control V group showed a 0.2 fall and control group NV a 0.4 rise in score, both statistically significant at p=0.05.
Contacts with GP and health workers was lower for the intervention group than for the control (V) in each of the four time periods (p<0.01).
There was a significant decrease in in the number of patients in the intervention group storing their drugs inappropriately (p<0.01); no statistically significant decrease was seen in any of the control groups.
The proportion of patients in the intervention group hoarding drugs significantly decreased from 61% to 0 at the two weeks and one month visits (p<0.001).
Safety and adverse effectsNone reported.
Does the study answer the question?Patients in the intervention group had better compliance, better drug storage practices and a reduced tendency to hoard drugs, and required fewer GP consultations than patients in the control groups.
Effect due to factor in study?Yes.
Consistency of results with other studies?
Directly applicable to guideline population?Relevant.
Internal Validity
Bernsten C;Bjorkman I;Caramona M;Crealey G;Frokjaer B;Grundberger E;Gustafsson T;Henman M;Herborg H;Hughes C;McElnay J;Magner M;van Mil F;Schaeffer M;Silva S;Sondegaard B;Sturgess I;Tromp D;Vivero L;Winterstein A;
Improving the well-being of elderly patients via community pharmacy-based provision of pharmaceutical care
Ref ID 179832008
Study TypeRandomised Controlled TrialFundingEuropean Commission funding.
Number of participantA total of 1290 intervention patients and 1164 control patients were recruited.
Inclusion/Exclusion CriteriaPatients were 65 years or older, taking 4 or more prescribed medications and oriented with respect to self, time and place. They were community dwelling and regular visitors to a recruited community pharmacy.
Patients were excluded if they were housebound or resident in a nursing/residential home. Identification of patients was performed via a personal approach by the pharmacy.
Patient CharacteristicsMedian age was 74 (s.d=8) for the intervention and control group. 42.1% were male and 57.9% were female in the intervention group. 42.7% were male and 57.3% were female for the control group.
RecruitmentStudy sites were selected using the responses of community pharmacists who expressed interest in participating in the research, following publicity via mailshots, advertisments in pharmaceutical publications and at professional meetings.
SettingCommunity pharmacies.
Interventions/Test/Factor being investigatedPharmaceutical care program by the trained pharmacists compared to usual care which was normal services provided to the recruited patients.
Pharmacy interventions included: 1) educating the patient about their drug regimen and their condition; 2) implementing compliance-improving interventions such as drug reminder charts; 3) rationalising and simplifying drug regimens in collaboration with the patients GP. This was a continuous process throughout the 18 months of the study.
ComparisonsBetween treatments.
Length of Study/Follow-upUp to 18 months.
Outcome measures studiedHospitalisations, quality of life, satisfaction with service provided, clinical signs and symptom control, knowledge of medicines, contact with GPs, prescription and nonprescription drug use.
ResultsSeven countries were involved: Denmark, Germany, The Netherlands, Northern Ireland, Portugal, Republic of Ireland, and Sweden. Drop-outs were higher in some countries that others, however most withdrew in the first 6 months. Those who withdrew from the study were significantly older (p<0.05) and reported poorer quality of life at baseline (p<0.05).
Generally, the programme had some positive effects on humanistic health outcomes such as satisfaction with treatment, and sign and symptom control, and on economic outcomes, but had less impact than anticipated on drug therapy, drug knowledge and compliance with medication.
An analysis of changes in compliance during the study indicated that at 18 months a significantly higher proportion of the intervention patients changed from being noncompliant to compliant compared with the control groups (p=0.028).
Intervention patients rated the services provided higher that the control at 6 and 18 months (p<0.05). There was a small statistically significant increase in satisfaction in the intervention group over time (baseline vs 12 months p=0.039).
Safety and adverse effectsNone.
Does the study answer the question?It is a large-scale multicentre study that assessed the effects of a pharmaceutical care programme by community pharmacists to elderly. Intervention patients reported better control of their conditions. The new service was well accepted by the intervention patients and patient satisfaction with the services improved during the study.
Effect due to factor in study?Yes.
Consistency of results with other studies?
Directly applicable to guideline population?Relevant.
Internal Validity
Hanlon JT;Weinberger M;Samsa GP;Schmader KE;Uttech KM;Lewis IK;Cowper PA;Landsman PB;Cohen HJ;Feussner JR;
A randomized, controlled trial of a clinical pharmacist intervention to improve inappropriate prescribing in elderly outpatients with polypharmacy
Ref ID 50121996
Study TypeRandomised Controlled TrialFundingGrant from the National Institute on Aging; An Academic Award from the National Institute on Aging; The Claude D. Pepper Older Americans Independence Center.
Number of participant208 in total were randomised, 105 to the intervention group and 103 in the control group.
Inclusion/Exclusion CriteriaInclusion: 65 years or over, evidence of polypharmacy (5+ medicines prescribed), received primary care in the GMC.
Exclusion: Residents of a nursing home, cognitively impaired (mental status questionnaire) were excluded unless a caregiver was available for involvement in intervention.
Patient CharacteristicsMean values: Mostly male 99%, white 77%, 70 years old, married (65.7% intervention, 85.4% control), compliance rates of 73.5%, medication knowledge 80.5%, 10 years of education, 9 chronic medical conditions, 8 prescribed medications, 3 medications recommended.
RecruitmentThose with regular scheduled medications by a Veterans Affairs physician receiving primary care in a General Medicine Clinic; computerized and manual chart audits identified participants.
SettingThe Durham Veterans Affairs Medical Centre GMC.
Interventions/Test/Factor being investigatedUsual care plus pharmacist intervention. Before the patients visit to the GMC the clinical pharmacist monitored their drug therapy outcomes by reviewing their medical records and medication lists and ascertaining their current medication use, drug- related problems and evaluating their needs by applying the Medication Appropriateness Index. This was then reported to the physician. After the visit to the physician the pharmacist educated the patient on the drug-related problems and encouraged compliance with strategies such as medication reminder packages or calendars and written patient materials. Reviewed principles of safe medicine use and the importance of discussing medications with their physicians.
ComparisonsPharmacist intervention versus usual care (which included a clinical nurse reviewing patients current medications before their visit, the physician visit and then the nurse reviewing and medication modifications).
Length of Study/Follow-upFollowed up for one year (Last telephone interview between 11.5 to 13 months after randomisation).
Outcome measures studiedPrescribing appropriateness; Health-related quality of life; Potential adverse drug events that had occurred during the past year; Patient compliance and knowledge; Patient satisfaction at end of year.
ResultsCompliance was assessed by patient self-report. There were no significant differences between the groups at the end of the follow-up period with regard to medication compliance (77.4% of intervention group and 76.1% of control group complied, p=0.88) knowledge, number of medications or patient health care satisfaction.
More control patients experienced adverse drug events than the intervention group (40% vs 30.2%, p=0.19).
Written recommendations were enacted more (by physicians) in the intervention group than the control group (55.1% vs 19.8%, p<0.001).
Safety and adverse effectsNone reported.
Does the study answer the question?It does partially, however it should be noted that the pharmacist intervention involves not only medication review but medication education and compliance strategies.
The study did not find that these increased compliance to medication, therefore this suggests that an intervention which included pharmacist medication review did not have an effect on compliance to medication.
Effect due to factor in study?Yes
Consistency of results with other studies?
Directly applicable to guideline population?Patient population is of interest for this guideline the intervention is partially comparable to the intervention of interest.
Internal ValiditySubjects not blinded to treatment.
Nazareth I;Burton A;Shulman S;Smith P;Haines A;Timberal H;
A pharmacy discharge plan for hospitalized elderly patients-a randomized controlled trial
Ref ID 74842001
Study TypeRandomised Controlled TrialFundingThe National Health Service research and development programme.
Number of participant362 patients, 181 to the intervention and control group.
Inclusion/Exclusion CriteriaInclusion: over 75 years and taking four or more medicines at discharge and living in the hospitals catchment area.
Exclusion: not speaking English or too ill.
Patient CharacteristicsMean age of participants 84 years in both intervention and control group (s.d=5.2 and 5.4 respectively).
62% of intervention and 66% of control group were women. 97% were white. Each patient had a mean of three chronic medical conditions and on mean 6 drugs (s.d=2).
RecruitmentPatients discharged from three acute general and one long-stay hospital in a health authority in central London.
SettingCommunity pharmacists visited at home.
Interventions/Test/Factor being investigatedPharmacist check for discrepancies with the medicine taken and those prescribed. Assessing understanding and adherence to the medication regimen and intervened when appropriate. Counselling patients/carers on correct dosage, disposing of excess medicines and liaising with gps.
ComparisonsIntervention vs control group - who were discharged with standard procedures - a discharge letter to the gp indicating the diagnosis, investigations and current medications, no pharmacist review of medication or follow-up.
Length of Study/Follow-upAt 3 and 6 months.
Outcome measures studiedPrimary outcomes: re-admission to hospital in follow-up period.
Secondary outcomes: number of deaths, attendances at hospital outpatient clinics and gps. well-being, satisfaction with service, adherence to and knowledge of medication, hoarding of meds.
ResultsThere was no significant differences in any of the outcome scores except patient knowledge.
There was no significant difference in the mean adherence scores of those re- admitted to hospital and the rest of the subjects at 3 and 6 months.
At 3 months: adherence to medicines: 79 (52%) mean 0.75 (s.d=0.3) in the intervention group and 72 (48%) mean 0.75 (s.d=0.28) for the control group. 95% CI 0.
At 6 months: adherence to medicines: 60 (45%) mean 0.78 (s.d=0.3) in the intervention group and 58 (43%) mean 0.78 (s.d-0.3) in the control group. 95% CI 0.
Safety and adverse effectsNone.
Does the study answer the question?Yes. Adherence to medication did not increase from a pharmacy discharge intervention with elderly patients.
Effect due to factor in study?The methodology was adequately addressed apart from blinding was not reported and they did not recruit to the statistical power they required. Therefore it is unsure that the effect is due to the intervention.
Consistency of results with other studies?
Directly applicable to guideline population?Yes.
Internal ValidityBlinding
Sturgess IK;McElnay JC;Hughes CM;Crealey G;
Community pharmacy based provision of pharmaceutical care to older patients
Ref ID 24882003
Study TypeRandomised Controlled TrialFundingSupported by (no details of type of support given) Northern Pharmacies Trust, Northern Ireland and European Commission under the BIOMED 2 programme.
Number of participantTotal sample: 191 patients. Intervention group: 110, Control group: 81.
Inclusion/Exclusion CriteriaInclusion: elderly patients (? 65 years) who were community dwelling, taking four or more prescribed medications, regular visitors to the participating community pharmacy and orientated to self, time and place were eligible.
Exclusion: Patients were excluded if they were housebound or living in a nursing/residential home.
Patient CharacteristicsAge (years): intervention group: 73.1 (s.d=5), control group: 74.2 (s.d=6.3). Gender (% male/% females): intervention group: 36.4/63.6, control group: 39.0/61.0. There were some differences between the two groups at baseline in mean number of prescribed medications (higher in control group, p=0.05) and SF-36 domains of mental health (intervention group higher score, p=0.05), physical functioning (intervention group higher score, p=0.05) and vitality (intervention group higher score, p=0.05).
Setting10 pharmacies in Northern Ireland.
Interventions/Test/Factor being investigatedNote: Only half of the sites saw the project through to completion (3 intervention (from five randomised to deliver intervention) and 2 control (also from 5 original)).
Delivered by community pharmacists. Intervention pharmacists assessed patients to identify drug-related problems. A number of information sources were used by intervention pharmacists during this assessment procedure including: the patient (via informal questioning), the patient’s gp, study questionnaires and computerised medication records. During the assessment, pharmacists were asked to document any identified drug-related problems and to form with the patient an intervention and monitoring plan e.g. education, implementation of adherence improving strategies. Pharmacists visited patients at home to assess storage of medicines where problems were identified.
ComparisonsPharmaceutical care programme (PCP) (intervention) v usual care. Intervention vs Control.
Length of Study/Follow-up18 months.
Outcome measures studiedPrecise Items used to measure adherence not given (given in a separate publication) although self report scale and refill compliance rates are reported in the analysis. All measurements taken at 6, 12 and 18 months.
ResultsAdherence: Self reported compliance: between-group analysis at each assessment point indicated that a significantly higher proportion of intervention patients were compliant with their medicine at 12 (intervention group: 40.4%, control group: 24.4%) and 18 (intervention group: 47.3%, control group: 14.7%) months compared to control patients (p<0.05) (6 months: intervention group: 34.5%, control group: 29.4%). Analysis of change in compliance during the study (change in compliance status compared to that reported at baseline) showed that a significantly higher proportion of intervention patients changed from non-compliant to compliant compared to control patients (intervention 13.4% vs control 9.1%) and a significantly higher proportion of control patients changed from compliant to non-compliant compared to intervention patients at 18 months (control 36.4% vs intervention 4.5%). Refill compliance results: between-group analysis at each assessment point indicated that a significantly higher proportion of intervention patients were compliant with their medicines at six months (intervention group: 46.2%, control group: 19.1%) compared to control patients (p = 0.02) (results 12 months: intervention group: 40.4%, control group: 25.0%. 18 months: intervention group: 40.0%, control group: 40.6%). Analysis of change in compliance during the study (change in compliance status compared to that reported at baseline) showed no differences between control and intervention patients.
Other outcomes: Health related quality of life: During the study there was a trend for intervention patients’ quality of life to decline over the 18 months whilst that of control patients appeared to significantly improve in some of the SF-36 dimensions (physical functioning: intervention group change: ?6.83, control group: +7.14 and vitality, intervention group change: ?2.26, control group: +7.24, p<0.05), however, these findings were largely driven by patients attending one control site pharmacy who showed marked improvements in SF-36 scores over time. There was no significant difference between the two groups in terms of the number of hospitalizations, the extent of prescription drug use (after baseline) and knowledge about medications. Longitudinal analysis indicated that intervention patients were taking significantly more prescribed medicines at 6 (6.13, s.d=2.32), 12 (6.63, s.d=2.72) and 18 months (6.20, s.d=2.32) compared to baseline (5.87, s.d=1.86; p<0.05), whilst that of control patients remained constant. Problems with medications: There were no significant differences between control and intervention patients during the first 12 months of the study, however, during the last 6 months, intervention patients (0.90, s.d=1.27) reported significantly fewer problems with their medicines compared to control patients (2.09, s.d=2.38) (p<0.05). There were no differences between the two groups in their reported contact with nurses, however, there were differences in GP contacts and contact with a specialist during the study. Intervention patients reported higher numbers of contacts with their GP during the first (0–6) (2.89, s.d=4.44) and second (7–12) (2.97, s.d=2.56) six month periods than control patients (0–6: 1.88, s.d=2.55. 6–12: 1.97, s.d=4.25) (p<0.05). In addition, intervention patients reported more contact with a specialist during the second (7–12) (0.89, s.d=1.25) and third (13–18) (0.87, s.d=2.60) six-monthly periods compared to control patients (7–12: 0.16, 0.50. 13–18: 0.10, s.d=0.31) (p<0.05).
Safety and adverse effectsNone.
Does the study answer the question?Yes. The intervention helped to increase adherence according to the majority of analysis undertaken.
Effect due to factor in study?Fairly. Baseline differences between groups a potential confounding factor.
Consistency of results with other studies?
Directly applicable to guideline population?Relevant.
Internal Validity
Zermansky AG;Petty DR;Raynor DK;Lowe CJ;Freemantle N;Vail A;
Clinical medication review by a pharmacist of patients on repeat prescriptions in general practice: A randomised controlled trial
Ref ID 75442002
Study TypeRandomised Controlled TrialFundingHealth Technology Assessment Programme.
Number of participant1188 in total. 608 in the intervention group, 590 in the control group.
Inclusion/Exclusion CriteriaInclusion: 65 years or older on repeat medication.
Exclusion: in a clinical trial, a residential or nursing home or having a terminal illness.
Patient CharacteristicsData not found for ethnicity but the study was mainly a Caucasian population born in the UK.
RecruitmentA note was attached to their last prescription before their due date. This said to book an appointment with the practice receptionist.
SettingLeeds gp practices with 4 or more partners.
Interventions/Test/Factor being investigatedPharmacist medication review to make recommendations on medication changes.
ComparisonsBetween intervention and control group.
Length of Study/Follow-up12 months.
Outcome measures studiedPrimary outcome - number of repeat medication changes for each patient.
Secondary outcomes - effect on the medication costs; whether medication review taken place (intervention group vs control group).
ResultsThe mean number of individual medication changes per patient were 2.2 intervention group vs 1.9 in control group (0.31, 95% CI 0.06 to 0.57, p=0.02.
The number of repeat items rose in both groups but was significantly less for intervention group (0.2 mean, SD 1.55), control (0.4, s.d=1.53, difference −0.2, 95% CI −0.4 to −0.1).
Medication costs rose in both groups but the rise was significantly less in the intervention group £1.80 mean compared to £6.53 mean for control group, difference was £4.75 per 28-day month. Saving of £61.75 per patient per year.
97% of intervention group had medication reviews compared with 44% of the control group.
The most common recommendation was to stop the medicine or removal of a redundant item from a list.
Safety and adverse effectsNone.
Does the study answer the question?It helps answer about the effectiveness of medication review but adherence is not a main outcome measured.
Therefore it will be included in the introduction for medication review but not as an evidence narrative on medication review increasing adherence.
Effect due to factor in study?Yes.
Consistency of results with other studies?
Directly applicable to guideline population?Intervention very relevant for guideline but not adherence outcomes.
Internal ValidityNo blinding
Grading: 1−Meta-analyses, systematic reviews of RCTs, or RCTs with a high risk of bias*
Chisholm MA;Mulloy LL;Jagadeesan M;DiPiro JT;
Impact of clinical pharmacy services on renal transplant patients’ compliance with immunosuppressive medications
Ref ID 612001
Study TypeRandomised Controlled TrialFundingGrant from the Carlose and Marguerite Mason Trust Fund.
Number of participant24 in total. 12 in the intervention group and 12 in the control group.
Inclusion/Exclusion CriteriaInclusion criteria: Aged 18 to 60 years; had only one kidney transplant; received follow-up care at MCG for at least one year after transplant; prescribed same immunosuppressant for at least one years since transplant; received immunosuppressant from MCG Outpatient Pharmacy for whole year.
Patient Characteristics75% were male, and 58.3% Caucasian, 37.5% African-American and 1 Hispanic. 33% had living-related donor kidneys, 67% had cadaveric kidneys. The mean age was 49 (s.d=10.2). Twenty one of patients prescribed cyclosporine and the other 3 had tacrolimus.
RecruitmentAll patients who had a renal transplant at MCG from February 1997 to January 1999.
SettingMedical College of Georgia Hospital and Clinics.
Interventions/Test/Factor being investigatedIn addition to usual care, patients received direct patient care clinical services from a clinical pharmacist. They obtained medication histories and reviewed medications with emphasis on optimising medication therapy to achieve desired outcomes and to minimse adverse events. They also made recommendations to the nephrologists to get the desired outcomes. The pharmacists counselled patients on ther medication and instructed how to take correctly (verbally and/or in writing). The patients were encouraged to call the pharmacist with any questions or concerns. The patients understanding of their medication was assessed. The medication reviews and histories were conducted monthly for the intervention group. Compliance enhancement principles were used at visits or by phone.
ComparisonsBetween the intervention group and the control group who received usual care but had no clinical pharmacist interaction.
Length of Study/Follow-up12 months.
Outcome measures studiedCompliance rate, directly observed by immunosuppressive serum concentrations.
ResultsAt end of 12 months the mean compliance rate was 96.1% (s.d=4.7%) for the intervention group and 81.6% (s.d=11.5%) for control group, p<0.001 statistically significant. For 6 of the 12 months 6–8 and 10–12) there were differences in compliance rates (64–100% for control group and 89 to 100% for intervention group) always with the intervention group higher rates (p<0.05).
Duration of compliance differed also, with the intervention group remaining 75% compliant each month whereas only 33.3% of the control group remained compliant (p<0.05).
Intervention patients had a greater achievement of ‘target’ serum concentrations than control patients (p<0.05).
Safety and adverse effectsNot mentioned.
Does the study answer the question?Yes. Patients who received clinical pharmacy services along with routine traditional patient care services had better immunosuppressive compliance than patients who only received traditional patient care services. The mean compliance rate for intervention was higher than the mean for the control group. Those in the intervention achieved higher achievement of the target immunosuppressive serum concentrations than the control group.
The pharmacist intervention is beneficial for enhancing medication compliance in post-transplant patients.
Effect due to factor in study?The study was very small, with only 24 participants and the methodology was not very strong so it can not be certain that the effect is due to the study intervention. Although all measurements were consistently higher for the intervention than the control group.
Consistency of results with other studies?
Directly applicable to guideline population?Not only medication review but includes counselling, compliance-enhancing techniques. Not generic medication review.
Internal ValiditySelection bias; performance bias; small sample;
Grymonpre RE;Williamson DA;Montgomery PR;
Impact of a pharmaceutical care model for non-institutionalised elderly: Results of a randomised, controlled trial
Ref ID 21752001
Study TypeRandomised Controlled TrialFundingNot mentioned. Authors are from a University and one was a pharmacy consultant.
Number of participant135 in total, 69 in the intervention group and 66 in the control group.
Inclusion/Exclusion CriteriaInclusion criteria: 65 years or over, non-institutionalised, taking two or more prescribed or non-prescribed medications, and providing signed consent form.
Patient CharacteristicsMostly female (75% intervention vs 83% control, p=0.254); aged 76.9 (s.d=8.4) and 77.2 (s.d=8.8), p=0.786. All were Caucasian, Most lived alone 61% vs 77%, p=0.018)
RecruitmentClients who presented at a clinic or were referred by Home Care programme.
SettingA community-based health clinic.
Interventions/Test/Factor being investigatedVolunteers and staff were trained to conduct a comprehensive medication review and this is given to the pharmacist to identify and document potential and actual drug- related issues and to address the issues with the patient and their physician. This included their use of prescribed and non-prescribed medicines, social drugs, home remedies, their regime, their adherence and their communication with, any problems or side effects with drugs. The recommendations were given in a letter to physicians and were reviewed for appropriateness by a consultant geriatrician before given to the physician. The clients were followed up by the pharmacist when required to monitor therapeutic endpoints and sort out any problems that had arisen. The issues identified by the pharmacist were tested individually by a pharmacist and nurse to see if resolved. Physicians gave their opinion of the pharmacist’s letter through a survey.
ComparisonsBetween intervention group and control group. The control group received a detailed home medication history but were reviewed by a different pharmacist who referred clients to their usual pharmacist and answered any queries.
Length of Study/Follow-upNo data given.
Outcome measures studiedNumber of drugs taken, drug knowledge, adherence to drug therapy, cost of prescribed medicines, number of symptoms reported from home medication history, response of physicians’ survey.
ResultsThe mean number of mediations adhered to at follow-up was 87 (+/−46) for the intervention and 85 (+/−41) for the control group, p=0.895, showing no significant difference in adherence.
Safety and adverse effectsIf the pharmacist thought the clients were at risk of ‘life-threatening’ drug-related problems in the control group they were withdrawn from the study.
Does the study answer the question?Yes. A medication review and recommendations given by the pharmacist to physicians did not change adherence or drug knowledge between the intervention and control group.
Effect due to factor in study?The methodology is lacking in that the two groups may have been treated similarly and so a difference between the two groups would not be evident.
Consistency of results with other studies?
Directly applicable to guideline population?The intervention is comparable to the intervention and population of interest as it is medication review and measures adherence. However the medication history collection is conducted by a lay person rather than the pharmacist (who conducts the review).
Internal ValidityAttrition bias; Not blinded; group contamination.
Lipton HL;Bird JA;
The impact of clinical pharmacists’ consultations on geriatric patients’ compliance and medical care use: a randomized controlled trial
Ref ID 16271994
Study TypeRandomised Controlled TrialFundingJohn A Hartford Foundation in New York City.
Number of participant1,383 eligible patients approached, 10% refused, 37% discharged before deciding whether or not to enrol.
52% of patients who were eligible and approachable were enrolled. After attrition (6.5%) 706 patients remained in the trial.
Inclusion/Exclusion CriteriaInclusion: aged 65 years or over; covered by Medicare; admitted to a non-psychiatric ward; resided within 35 miles; English speaking (or proxy); mental competent (or proxy); access to telephone; 3/4 medications prescribed for a chronic condition;
Exclusion: those discharged to a nursing home or hospice;
Patient CharacteristicsIntervention vs control groups:
Mean age: 74 both groups
MediCal recipients: 9% both groups
More than 12 years education: 52% vs 44% (p=0.03)
All patients were discharged from hospitals. No mention of sex, ethnicity, comorbidity, disease status given.
RecruitmentDaily hospital records were looked at for eligible patients. At least one attempt was made to approach every patient meeting the eligibility criteria.
SettingCommunity hospital in San Francisco Bay, USA.
Interventions/Test/Factor being investigatedTwo clinical pharmacists’ provided a drug consultation service for geriatric patients and their physicians.
Intervention: Pharmacists’ reviews of the hospital records and drug regimens of the experimental, and consultations with the patients and their physicians.
Both control group and experimental group patients were given booklets when discharging from hospital, to record medication information eg drug purpose, dosage and schedule. After review of the records to determine the patient’s (in intervention group) clinical condition and to assess appropriateness of prescribing, the pharmacist conducted a face-to-face consultation with the intervention patients to discuss the purpose and use of their medications and any potential drug-related problems.
Follow-up was about 15 minutes in duration. 85% of the postdischarge meetings were by telephone and the rest were in the pharmacists’ office or patient’s home. If significant problems were detected the patients were provided with a consultation with their physician.
The pharmacists promoted the use of fewer medications and simplified regimens where appropriate – by telephoning physician to recommend discontinuation of a prescribed product or by recommending directly to the patient discontinuation of a non-prescribed product.
Patient compliance was assessed by structured telephone interviews with a subsample of experimental and control patients at 6–8 weeks postdischarge and again at 12–14 weeks postdischarge.
ComparisonsIntervention vs usual care.
Length of Study/Follow-upFollow-up consultations were given at 1 week, 2–4 weeks, 2 months and 3 months after discharge from hospital. 6 months.
Outcome measures studiedMedical care utilisation; Patient compliance; Knowledge, regularity, frequency, dosage, missed doses; polypharmacy.
ResultsT-test results showed that the intervention did not have an impact on subsequent medical care utilisation and expenditures.
No significant differences found for the mean number of drugs taken and the complexity of the regime at 6–8 weeks but there was a significant change at 12–14 weeks. Intervention group were taking significantly fewer medications than controls (5.16 vs 6.75, p<0.001). The intervention also had an impact on the second measure of regimen complexity, average daily doses per drug (p=0.02).
Compliance results: 274 patients were selected for this sub-study. No significant demographic differences between this sample and the overall sample were found. 233 (124 intervention and 109 control) were interviewed for the first assessment and 206 (108; 98) for the second assessment. During the first assessment (6–8 weeks) intervention group had significantly higher mean compliance 94.4 (s.d=9.4) vs 91.4 (s.d=11.6) (p=0.035). This became non-significant (p=0.334) when knowledge was removed from the analysis.
At 2nd assessment the interventions impact on knowledge was stronger (p=0.001). By this time the intervention had an effect on patients’ drug use 96.3 (s.d=10.2) vs 91.2 (s.d=9.6) (p<0.001). With 92% of intervention vs 77% of control patients not missing any dose of their medications (p<0.001). This was still significant whether or not knowledge of the purpose of the medication was included.
Safety and adverse effectsNone reported.
Does the study answer the question?Clinical pharmacist’s consultations can improve geriatric patients’ drug regimens and compliance. The need for replication among large cohorts of patients at high risk.
Shows the value of sustaining the clinical pharmacist intervention for some time.
Effect due to factor in study?No.
Consistency of results with other studies?
Directly applicable to guideline population?Yes
Internal ValidityAllocation concealment. Difference in the group.
Lowe CJ;Raynor DK;Purvis J;Farrin A;Hudson J;
Effects of a medicine review and education programme for older people in general practice
Ref ID 75372000
Study TypeRandomised Controlled TrialFundingGrant from the Department of Health under the Pharmacy Practice Research Enterprise Scheme.
Number of participant161 patients in total: 77 in the intervention group and 84 in the control group.
Inclusion/Exclusion CriteriaInclusion criteria: 65 years or older; taking 3 or more drugs.
Exclusion criteria: lived in nursing or residential care; dependent on another to administer medicine; terminal illness with life expectancy less than one year.
Patient CharacteristicsIntervention group: mean age 77.5 (65–96), mainly female 67%, living with spouse or relative 55% and 4 mean medicines scheduled (2–8).
Control group: mean age 75 (65–88), 67% female, 57% living with spouse or relative, 4% mean (1–10) medicines scheduled.
RecruitmentThey were recruited sequentially from a list of patients in the practice 65 or over.
SettingGeneral practice in suburbs of Leeds.
Interventions/ Test/ Factor being investigatedAn investigator visited intervention and control participants and filled in a structured questionnaire regarding their medicines, medicines taken and understanding of their purpose. The investigator assessed the intervention group participants' ability to take their medications, then reported the findings to doctors where there was need to reduce dosage and discontinue medication. They also liaised with pharmacist for modifications to medicine containers.
At the second visit they gave 1 months supply of medication and removed any other prescribed medications. They discussed the regimen and explained the right way to take medications and purpose and made a reminder chart. At 3 weeks follow-up another months supply was given and the patients were asked to describe the medicines they took and their purpose, and the medications left over from the last visit were counted.
ComparisonsComparison made between intervention group and control group - who did not receive the intervention of medication review, education and discussing medication and problems.
Length of Study/ Follow-upFollowed up after one month, then after 3 weeks.
Outcome measures studiedKnowledge of medicines, compliance with medicines - through a structured questionnaire and tablet count and patient report.
ResultsThe mean compliance score was 91.3% for intervention group (95% CI 89% to 94%) and 79.5% for the control group (95% CI 75% to 84%), p<0.0001.
At first visit 58% of intervention group correctly described the purpose of medication, compared to 67% of control these numbers were 88% of intervention and 70% of control group by the third visit, between groups the difference was significant (p=0.0001).
47% of patients had a fall in the mean number of medicines to take from 4.1 (95% CI 3.8–4.5) to 3.9 (95% CI 3.5 to 4.2) the mean difference was −0.26 (95% CI p=0.003).
Safety and adverse effectsApproval given by Local Research Ethics Committee and informed consent from patients.
Does the study answer the question?Yes this does answer the key question. The use of a medicine review and education increased compliance for the intervention group compared to the control group.
Effect due to factor in study?Uncertain as to whether there may have been bias introduced into the study. The statistical power of the study was high. The overall effect is possibly due to the study intervention.
Consistency of results with other studies?
Directly applicable to guideline population?Intervention is under 6 months so is not exactly the requirement for the guideline but the intervention involves medication review as the intervention and compliance as an outcome so this is of direct interest to guideline.
Internal ValiditySelection bias, performance bias
Sookaneknun P;Richards RM;Sanguansermsri J;Teerasut C;
Pharmacist involvement in primary care improves hypertensive patient clinical outcomes
Ref ID 15922004
Study TypeRandomised Controlled TrialFundingResearch grand from Chiang Mai University, Thailand.
Number of participant235 total patients: 118 in treatment group, 117 in control group.
Inclusion/Exclusion CriteriaInclusion: over 18 years; newly diagnosed during the pre-test period with hypertension; average DBP over or equal to 90 mm Hg; or average SBP over or equal to 140 mm Hg
Exclusion: secondary causes of hypertension; unable/unwilling to return for appointments; planned to move/family member in study; SBP over 210 mmHg or DBP over 115mg Hg; severe complicating disease.
Patient Characteristics76 women and 42 men in the treatment group; 84 women and 33 men in the control group p value 0.224; aged 63 (s.d=9), p=0.982; hypertension 57 vs 54; Hypertension with diabetes 39 vs 45; hypertension with target organ damage 13 vs 7; hypertension with diabetes and target organ damage 9 vs 11; p=0.474.
RecruitmentDatabases from hospital and 2 PCUs screened for patients diagnosed as hypertensive. Or from medical records.
SettingMahasarakham Uni community pharmacy, Thailand
Interventions/ Test/ Factor being investigatedPharmaceutical intervention:
30–50 minute face to face interview - assessed understanding of medications, counselled on use of medications, assessed adherence and lifestyle habits, reviewed for adverse events due to DRPs; identified, resolved and prevented DRPs; Pharmacist recommendations for regimen changes made to physicians and on medical record; also looked at lifestyle eg exercise; education leaflets and diary to record lifestyle presented.
ComparisonsPharmacist intervention versus usual care (no pharmacist involvement).
Length of Study/ Follow-up6 months.
Outcome measures studiedPrimary outcomes: Blood pressure control, blood pressure difference.
Secondary outcomes: adherence.
ResultsPrimary outcomes: significant reduction in both systolic and diastolic BP compared with the control group (p=0.037, 0.027, respectively).
Proportion of patients whose BP stabilised was higher in the treatment group (p=0.017).
Secondary outcome: the treatment group showed significantly better adherence 70% with good adherence in the treatment group compared to 60% of the control group and 40% showing poor adherence in intervention compared to 48% of control group (p=0.014) at the end of the study.
Safety and adverse effectsNone mentioned
Does the study answer the question?Yes.
Adherence was increased with the pharmacists involvement.
Effect due to factor in study?The study power was 90%, the target size of the study sample was 95 patients, with 30% added to allow for drop-outs.
Yes the effect is likely to be due to the study intervention.
Consistency of results with other studies?
Directly applicable to guideline population?Relevant as secondary outcome was change in adherence, from pharmacist involvement, which included medication review.
Internal ValidityRandomisation, concealment allocation.
Taylor CT;Byrd DC;Krueger K;
Improving primary care in rural Alabama with a pharmacy initiative
Ref ID 462003
Study TypeRandomised Controlled TrialFundingSupported by the ASHP Research and Education Foundation.
Number of participant69 in total, 33 in the intervention arm and 36 in the control arm.
Inclusion/Exclusion CriteriaAdults (over 18s) receiving care within the clinics. Those who were at high risk of medication-related adverse events (five or more medications prescribed, 12 or more doses per day, four or more medication changes in the last year, three or more concurrent diseases, previous medication compliance, drugs that require therapeutic monitoring).
Exclusion criteria: significant cognitive impairment, history of missing office visits, scheduling conflicts or life expectancy under a year.
Patient CharacteristicsMost patients were female 63.6% in the intervention group and 72.2% in the control group (p=0.445), Most were white 60.6% vs 61.1% (p=0.966), and mean age was 64.4 and 66.7 years respectively (p=0.467) and the majority were married 75.8% vs 72.2 (p=0.935) with 12 years mean education in both groups. They were attending community-based practices. Taking on average six medications each.
RecruitmentIdentified by pharmacist evaluation of clinic medical records (manual and computer) from physician's offices, of the three community-based family medicine clinics.
SettingGP offices, Alabama, USA.
Interventions/ Test/ Factor being investigatedFour pharmacists joined the clinics to give medication reviews. The intervention group received usual medical care, as did the control group but additionally received pharmaco-therapeutic interventions from a pharmacist during office visits.
The pharmacists purpose was to prevent or identify and resolve problems with drug therapy.
They evaluated a drug therapy's indication, effectiveness, and dosage as well as the correctness and practicality of directions, drug-drug interactions, drug-disease interactions, therapeutic duplication, the duration of treatment, untreated indications, and expense. They reviewed medial records for medication-related problems, documented problems accurately and examined medication history to determine compliance and complications with medication and gave individualised patient education reviewing the disease, lifestyle modifications and basic drug information. Therapeutic recommendations were made to the physicians and they made follow-up visits and gave more information or answered questions. Monitoring patients' responses to drugs and consolidating medication regimens, reducing dosage frequency, devising medication reminders and teaching techniques for using certain devices eg inhalers.
ComparisonsBetween intervention and no intervention.
Length of Study/ Follow-up12 months follow-up.
Outcome measures studiedClinical outcomes: Hospitalisations and emergency department visits, hypertension, diabetes mellitus, dysipidemia, anticoagulation, quality of life.
Prescribing appropriateness and medication misadventures: edication compliance and medication knowledge.
ResultsThe intervention group's percentage of patients with medication compliance scores of 80–100% increased by 15%, but there was no change for the control group. However there was no significant difference at 12 months between the groups (100% of patients in the intervention group versus 88.9 (s.d=6.3) of the control group had compliance scores of 80–100% at 12 months, p=0.115). At baseline this was 84.9% (s.d=6.7) and 88.9 (s.d=5.8) p=0.728 respectively.
The most frequently cited reasons were: forgetting to take the medications (n=10), having too many to take (n=9), finding it hard to read or understand the directions (n=4) and too much trouble (n=4).
Hospitalisations and Emergency Department visits decreased for the intervention group by 92% and 78% respectively, whereas the control group stayed constant. NB there was a much higher number of hospitalisations and ED visits in the intervention group than the control group at baseline 11 versus 24 hospitalisations and 6 versus 18 ED visits.
Safety and adverse effectsNot mentioned.
Does the study answer the question?Yes
There was increased compliance in the group who received the pharmacists' review of medications compared to the control group who received usual care. However this was not a significant difference in compliance at 12 months.
Effect due to factor in study?It is unclear as there is no time period or statistical power given for the result that there was increased compliance in the intervention group, but there is for twelve months, which was non-significant.
There was no concealment allocation so there may have been selection bias for the intervention group, although baseline scores were similar except for hospitalisation and ED admission which was higher in the intervention group, but then decreased significantly while the control group was constant.
Consistency of results with other studies?
Directly applicable to guideline population?Yes this intervention and population is directly comparable to those of interest for the guideline.
Internal ValiditySelection bias; self-reporting bias;

From: Appendix C, Clinical Evidence Extractions

Cover of Medicines Adherence
Medicines Adherence: Involving Patients in Decisions About Prescribed Medicines and Supporting Adherence [Internet].
NICE Clinical Guidelines, No. 76.
National Collaborating Centre for Primary Care (UK).
Copyright © 2009, Royal College of General Practitioners.

All rights reserved. No part of this publication may be reproduced in any form (including photocopying or storing it in any medium by electronic means and whether or not transiently or incidentally to some other use of this publication) without the written permission of the copyright owner. Applications for the copyright owner’s written permission to reproduce any part of this publication should be addressed to the publisher.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.