Hypothesis: Medication Review with follow-up can improve clinical, health related quality of life and economic outcomes. To prove this hypothesis a cluster randomized controlled trial will be held in primary care centres of the public health system of Chile. Patients of the cardiovascular disease prevention program, older than 65 years and with poly pharmacy (more than 5 drugs) will be recruited. Control group will receive usual care and the intervention arm will have medication review consultations by a pharmacist every 4 months for one year. Clinical interventions will be made with physician authorisation. Participating pharmacist will be trained in cardiovascular prevention pharmacotherapy in the elderly, interview skills and educational techniques. A practice change facilitator will assist the pharmacist in any matters regarding the methodology and will asses barriers and facilitators to the implementation of the medication review with follow-up service. A personalised plan will be developed for every pharmacist. Clinical outcomes (blood pressure, HbA1c, LDL cholesterol, overall cardiovascular risk, among others), number of medications, adherence rate and health-related quality of life will be evaluated. A cost-utility analysis will be made through the health ministry of Chile perspective.

This study was approved by the ethics committee from the Metropolitan Southeastern health service and the University of Technology Sydney (UTS)'s Human Research Ethics Committee (HREC). All patients are compelled to sign an informed consent document before the first interview, explaining that they can leave the study whenever they want without punishment or justification. All data will be coded and stored without any personal information, to comply with the Chilean law 19,628 for protection of personal data and 20,584 for rights and duties of the patients.

Sample size and losses

Sample size was calculated with data from the pilot study conducted between March and July of 2017. The effect size was determined by the reduction of CHD risk (0.324). Cluster size was of 24, clustering effect of 1.57, control-intervention relationship of 1:1, with an 80% of statistical power and type I error of 5%. A 20% attrition rate was assumed. With this data the calculated sample size was 576, with 288 patients and 11-12 clusters in each study group.

In addition, primary care centers will recruit participants following the proportion of older adults (OA) in the Cardiovascular disease care program (PSCV in spanish) of each one . If a center gathers more than 10% of the total population, that center must recruit 10% of the sample in that group.

Study structure

This study will have two stages: stage one is the preparation of the participating pharmacist in the intervention group and stage two is the development of the fieldwork.

Training of pharmacists in the intervention group will consist of 15 hours of topics developed in theory and practice, focusing on the resolution of clinical cases. The topics will be:

• Study methods.
• Pharmacist-physician and pharmacist-patient relationship.
• Measurement of vitals (blood pressure, cardiac frequency, weight and height determination, waist circumference, and capillary glycaemia).
• Communication abilities and health education.
• Geriatrics pharmacotherapy in hypertension, type 2 diabetes mellitus and dyslipidemias.

Control group will only be prepared in control of vitals and survey application.

Facilitator pharmacist

Facilitator Pharmacists (FAPHA) are pharmacists trained in MRF and implantation of professional pharmaceutical services. Its main objective is to conduct a systematic evaluation of implantation factors of the MRF service in the Polaris program, being positive -facilitators- or negative -barriers- and to support the pharmacists that provide the service in each individual center and municipality. Each Pharmacist will be supervised by a FAPHA, who will conduct weekly evaluations in the initial phase and monthly thereafter during the followup period.

Each FAPHA will be trained in:

• Medication review with follow-up.
• Pharmacotherapy of chronic diseases in the elderly.
• Analysis of implantation factors of a pharmaceutical service.
• Intervention strategies to overcome barriers and increase facilitators.

Main functions of the FAPHA are:

• To promote the collaborative work between each pharmacist and authorities of the Family health care center (CESFAM in spanish) or municipality..
• To detect causes of barriers and facilitators of the service in the primary care center.
• To increase and spread facilitators between primary care centers.
• To overcome barriers by implementing specific action plans in each primary care center.
• To establish the implementation of MRF service according to the Polaris method.
• To register the process and analysis in the corresponding data sheet.

Medication review with follow-up (MRF)

MRF will be conducted according to the Polaris methods developed for this study. Each participant will be interviewed at least seven times, and possibly more in the intervention group.

Service offer

The MRF service will be offered in two levels:

• Health Service and municipalities: the investigation team will introduce the program to each health service authorities and to each municipality (managers of each health care center and health authorities).
• CESFAM: the program will be presented to all health care professionals in each CESFAM by the pharmacist of that center. If needed, the assigned FAPHA can support this with the help of the pharmacist in chief.

Invitation to the service

Different methods can be used and will be compared to perform patient recruitment. This will be defined by each health care center's availability of resources. Each method requires that the participants of the study to go to each interview with all their medications.

The suggested methods are:

• Primary care center customer service: the invitation to participate in the study will be made by each center customer service. The investigator team will prepare a script for the administrative personal to use for recruiting and a list of patients to be recruited by phone call.
• Pharmacy unit: patients will be recruited in the pharmacy unit by each participant pharmacist. This will be conducted with spontaneous recruitment, at the moment of dispensation. In that instance, each participant should sign the informed consent document.
• Referral of other professionals: patients could be invited to the service through referral to the pharmacist if they meet the inclusion and exclusion criteria.

It is suggested to make the community aware of the service by making presentations of the program in different community instances, like the council of users of each primary care centre where the patients show their problems and participate in designing and proposing health practices and services.

Intervention group

Initial interview:

With the data obtained during the pilot study, this interview should last maximum 30 minutes. The following structure will be followed:

• Review of clinical records and pharmacy registry. This should be conducted previously by the pharmacist, and should be used to complete before the interview some of the patient information in the registry like health problems, prescriptions, laboratory tests and any other relevant information.
• During the first interview, pharmacist will apply the Morisky-green survey of four questions to determine medication adherence and the EuroQoL 5D-3L survey to determine perceived quality of life.
• Complete review of medications brought by the patient, including prescription and self-administered drugs. In addition, information about the use of each medication will be assessed while interviewing, like drug adherence, knowledge about the medication, reason of the prescription, signs or symptoms of Adverse Drug Reactions (ADR), among other information.
• Any other relevant information will be registered, and the pharmacist must perform a review of the interview to summarize the topics of interest and to receive any additional information that the patient considers relevant and has not been spoken previously.

Invitation to next interviews should be made at the end of every session and scheduled according their prescription refill, to increase adherence to the program and to confirm the appointment with the patient. All information gathered in the first interview will be registered in the MRF profile. After the initial analysis, patient will be classified in compensation or follow-up stage.

Compensation stage

All patients will be classified in this stage if they are not controlled in their Hypertension, Type 2 Diabetes Mellitus or Dyslipidemias, to assess the greater risk of cardiovascular complications. These patients will have a greater number of interviews and interventions, until they accomplish control of their diseases or the pharmacist decides that the problem can't be resolved in primary health care. The frequency of appointments will be the following:

• Uncontrolled Hypertension: interviews and interventions will be every 7-14 days until achievement of blood pressure goal. This will require close work with physicians and nurses. Once the goal has been reached, the patient will continue on the follow-up stage.
• Uncontrolled Type 2 Diabetes Mellitus: interviews and interventions will be every 7-14 days until achievement of the glycaemic target, controlled by fasting values and capillary glycaemia in the following 7 days. After an initial achievement normal values, the patient will have an appointment every 30 days. Only with confirmation of an HbA1c value in the bellow the accepted limit, the patient will continue on the follow-up stage.
• Uncontrolled Dyslipidemias: interviews and interventions every 4-6 weeks, until achievement of lipid goals. Once the goal has been reached, the patient will be derived to the follow-up stage.
• Special cases: patients in which a moderate or severe ADR is suspected, or those who present a pharmacological interaction defined as relevant by the literature, will receive interviews and interventions every 7-14 days until the issue has been tackled. After that, the patient will be derived to the follow-up stage.

Follow-up stage

All patients that are controlled in the previously described health problems will be classified in the follow-up stage, with interviews every four months.

Patients can be moved between both stages when necessary and will be constantly evaluated in their laboratory results, vitals and clinical signs and symptoms. Independently of the stage of the patient, all surveys must be completed every two months.

Study and evaluation phase

This phase consists of a comprehensive analysis of the medications used by the patient, reviewing aspects of necessity, efficacy and safety.

According to the data of the pilot study, this phase should last about 30 minutes. After an analysis of the pharmacotherapy of the patient, an action plan must be developed. If it requires a prescription change it must be discussed with a physician. If an educational plan is developed, it can be implemented directly. This process should last about 30 minutes, and it could be developed through a designated physician of the health center or through appointments with each physician in charge of each patient.

Adverse drug reactions (ADR)

A screening to identify ADR will be conducted in each interview by therapeutic group and expected ADR in the patient. In this study, ADR will be classified by the following:

Type according to the World Health Organization (WHO):

• A: Augmented, exaggerated effects from the pharmacological action of the drug (like hypotension with anti-hypertensive medication).
• B: Bizarre, ADR that can't be explained by the drugs mechanism of action (like allergic reactions or Steven-Johnson syndrome).
• C: Chronic, associated with a prolonged use of the drug (like benzodiazepine dependence).
• D: Delayed, like malignancy or teratogenic effects.
• E: End-use, associated with a residual effect after suspension of the drug (like rebound effect).

Causality through Naranjo's Algorithm:

• Definite
• Probable
• Possible
• Unlikely

Severity classified by the WHO:

• Mild: mild clinical manifestations that does not require a therapeutic intervention or suspension of the drug.
• Moderate: significant clinical manifestations that do not put in risk the life of the patient, but that require suspension of the drug or additional therapeutic interventions.
• Severe: important clinical manifestations that threatens the life of the patient and requires immediate suspension of the drug and urgent therapeutic interventions.

Each suspected ADR must be reported to the Chilean pharmacovigilance online system (RED-RAM) of the Public Health Institute of Chile.

Follow-up interviews

Follow-up interviews are defined as any interview after the first, and could be in compensation or follow-up stage. With the data from the pilot study, follow-up interviews should last about 20 minutes.

Registry of phase times

All phases should be timed and must be registered by the pharmacist to determine the duration of the process of MRF. The minimum timed activities should be:

• Previous review of the clinical records.
• Each interview.
• Time to register all data.
• Time spent while studying the patient and elaborating the action plan.
• Time of the meeting with the physician.

Laboratory exams will be considered as valid if they are not older than six weeks -three months to HbA1c-. Each pharmacist must manage the exams orders by their own primary care center policies.

Control group

Control group will receive usual care, with clinical attention of health professionals like nurses, physicians and dietitians, dispensation of medications from technicians in the pharmacy unit, and consultation of the pharmacist if required by the patient. If there is a detection of a medication issue with a patient, it will be reported to the health center managers so it could receive usual care. Each patient must have interviews every two months, with seven points of comparison with the intervention group. In each interview, pharmacist will register the patient's vitals, medications and health problems; medication adherence will be measured through the four questions Morisky-Green test and QoL by the EuroQoL 5D-3L tests. Time spent in the interview must be registered. According to the pilot study it should be about 15 minutes.

Economic evaluation

Study perspective The study will be analyzed from the Chilean Ministry of Health (MINSAL) perspective. We elected to use this, because CESFAM are public funded institutions and should develop their services following the public policies of MINSAL.

Therefore, all the intervention related costs that will be included should be the ones related to public budget, for example, cost of implementation of the service, cost of development the intervention and any other cost related with public funding.

Comparators MRF will be compared with usual care as defined above. Time horizon The costs and benefits will be evaluated for a year to avoid any season-related bias and to obtain significant changes in QoL, number of medical appointments, emergency unit visits and hospitalizations, among others.

Discount rate

Discount won't be applied because of the extension of the study (one year); therefore the discount rate will be 0%.

Choice of health outcomes

The benefits will be measured through QALY (Quality Adjusted Life Year) . This analysis is widely used because it reflects the personal perception of patient's QoL through survey considering also the quantity of life gained with those values. For the calculation of QALYs we will use the regression-adjusted area under curve (AUC) method to control for differences in the initial mean values.

This approach allows comparison with other health technologies that also use QALY as a measure of benefit.

Measurement of effectiveness

The measurement of effectiveness is described above. Measurement and valuation of preference based outcomes The QoL of the patients will be estimated using the survey EuroQoL 5D test. Patients will determinate their own QoL perception, both in a descriptive and a general system with a Visual Analog Scale (VAS). The descriptive system has five dimensions (mobility, personal care, daily activities, pain and anxiety/depression); each one has three levels of gravity. In this section of the survey, the patient must select the level of gravity according to his/her perception in that day. In the VAS system, the patient must define his/her personal health perception, ranging it in a scale between 0 and 100, being 0 death and 100 the best health status possible.

Estimation of resources and costs

The resources and cost will be evaluated along the year of study. In accordance with MINSAL perspective approach, we will evaluate the following:

• Initial investment:

  • Resource: pharmacists training time for the study.
  • Costs: value by hour spent in the training program (determined by the academic entity).

• Pharmacist´s working time (minutes), measured in Chilean pesos (CLP). • Resource: time spent in the service.

Time spent in each phase of the service:

• Initial interview.
• Study and evaluation.
• Intervention with the physician.
• Intervention with the patient.
• Follow-up interviews.
• Anything else. o Cost: Pharmacist salary.

Pharmacists are paid according to the Chilean law number 19.378, which refers and involves primary health care professionals, and determines different categories and levels for a full-time job (44 hours a week):

• Categories:
• A: Physicians, Pharmacists, Biochemists and Dentists.
• B: Other professionals like Nurses, Dietitians, and Physiotherapists among others.
• C: superior-level technicians.
• D: mid-level technicians.
• E: Administrative staff.
• F: cleaning staff and others.
• Each employee begins in level 15 with a fixed salary that can be increased by two requirements: time (every two years) and training -by doing courses in a related area-. When an employee meets these two requisites, advances to the upper level.
• Employees with diplomas get an increase on their wage. It varies depending on the duration of the diploma (between 5 and 15%).

Chilean government fixes the minimum wage every year, but allows each municipality to determinate their own salaries as long as they are higher than de minimum. Therefore, the wage of each municipality will be used for the analysis.

There are bonifications given by the municipalities for different responsibilities. This bonifications will not be considered because they don´t represent extra cost for the intervention.

• Medications:
• Resource: quantity of medications dispensed during the study time obtained from the records from each primary care center.
• Cost: prices of medications are determined according to the Chilean national supply center (CENABAST in Spanish).
• Emergency visits: registered the year before the study and during the intervention.
• Resource: Hospital urgency records, obtained from each hospital.
• Hospital admissions: registered the year before the study and during the intervention.
• Resource: number of hospital admissions, obtained from each hospital.
• Costs: MINSAL has fixed prices according to diagnosis called GRD (group-related diagnosis).

This information will be used to determine if the patient was admitted by a drug-related problem (DRP). The evaluation will be conducted by three physicians, specialists in internal medicine according to Malet-Larrea et.al methods. The following information of the patient will be provided to the evaluators:

• Age.
• Gender.
• Health problems.
• Status of the health problems.
• Prescribed medication.
• GRD.

The specialists in internal medicine will answer yes or no; if he/she considers that the cause of the hospitalization was a DRP. Two yes answers are required to establish a positive causality. Cohen´s kappa index will determine the inter-rater agreement and Fleiss´s kappa index will be used for reliability of agreement between all raters

• Facilitator Pharmacist
• Resource: time spent in transport between visits, planning and intervening with each pharmacist participant in the study and registration of the process.
• Cost: FAPHA costs will be determined by their own salaries.

Currency, price date, and conversion The costs and resources will be measured during 2017 and 2018. All the analysis will be conducted in CLP. Year 2019 will be used as baseline year.

The ICER (incremental cost-effectiveness ratio) will be determined for each scenario, discounting the cost and benefits of usual care from the intervention group according to the following formula: ICER = (Intervention Costs - Control Costs) / (QALY of intervention - QALY of control) Uncertainty will be determined through bootstrapping to evaluate the variability of the outcomes by conducting non-parametric estimations in 5000 different cases. Bootstrapping results will be visualized in a cost-effectiveness plot, and an acceptability curve for different QALY prices. All patients who complete the four interviews will be considered in the analysis.

Inclusion Criteria:

• Age 65 or older.
• Polypharmacy, defined as five or more chronic prescribed medications.
• Independent or independent at risk, classified by the chilean scale of autonomy for older adults (EFAM in spanish).
• Included in the Cardiovascular Primary Care Program.

Exclusion Criteria:

• Low Cardiovascular Disease Risk (CVDR).
• Participants of the pilot study.
• Risk of dependency or dependent by the EFAM.

Each participant will sign an informed consent, without knowing that there is another group of patients.