|
A pharmacy-based health promotion programme in hypertension: cost-benefit analysis |
Cote I, Gregoire J-P, Moisan J, Chabot I, Lacroix G |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology A pharmacy-based health promotion programme, developed to improve blood pressure control through activities aimed at improving the quality of prescribing and adherence to treatment, was evaluated. Decision-support computer software was developed to facilitate the implementation of the intervention program by pharmacists. Each time an exposed participant came to refill his or her antihypertensive medication, the computer program flagged him or her as a study participant. The pharmacist would then measure the participant's blood pressure and record it in the computer program. The software was programmed to classify participants according to whether or not their blood pressure was controlled, to assess adherence to drug treatment, and to suggest appropriate interventions.
Study population The study population comprised individuals residing in the Quebec City area and taking antihypertensive medication.
Setting The setting was the community. The economic study was conducted in Quebec, Canada.
Dates to which data relate The study was undertaken between October 1998 and December 1999. The price year was 1998.
Source of effectiveness data The effectiveness data were derived from a single study. A detailed description of this study, providing details of the implementation process and the clinical impact, was reported elsewhere (Chabot, see Other Publications of Related Interest.
Link between effectiveness and cost data The costing was undertaken prospectively on the same patient sample as that used in the effectiveness data.
Study sample No sample size appears to have been determined in the planning phase of the study and no power calculations were performed retrospectively. From the list of pharmacies using drug distribution software compatible with the decision-support software to be used in the study, the authors selected ten pharmacies in the Quebec City area known to have a practice oriented toward the delivery of pharmaceutical care. Nine of these pharmacies agreed to participate. Pharmacists from four different outlets agreed to administer the health promotion programme to their patients for 9 months. The programme was not administered to patients attending the five other outlets.
A total of 111 individuals agreed to participate in the study. The authors reported that 11 individuals did not complete the study (5 exposed to the programme and 6 not exposed). Of these 11, one died during the intervention period, 6 could not be reached and 4 were not interested in continuing. Consequently, the study was completed by 100 participants of whom 41 were exposed to the intervention and 59 were not. Fifteen (36.6%) individuals in the exposed group were younger than 65 years old, compared with 30 (50.9%) individuals in the non-exposed group. Thirteen (31.7%) individuals in the exposed group and 22 (37.3%) in the non-exposed group were male.
Study design This was a quasi-experimental cohort study conducted in nine pharmacies across Quebec City area. The groups were followed for 9 months. Eleven patients did not complete the study (5 exposed to the programme and 6 not exposed). Each participant was visited prior to the intervention period (between October 1998 and March 1999) and following the intervention (between October 1999 and December 1999) by a research assistant who administered a computer-assisted structured questionnaire. The research assistant was blinded to the exposure state of the participants.
Analysis of effectiveness The analysis was conducted on the basis of treatment completers only. The primary outcomes used were the mean after-before changes in systolic and diastolic blood pressure. Patients enrolled in the nine pharmacies had their blood pressure measured at home, both at baseline and 9 months later, to compare mean after-before changes in systolic and diastolic blood pressure between the two groups. The participant characteristics were generally statistically similar across the two groups. There were three exceptions. The proportion of participants using a car to go to the pharmacy or the clinic was lower in the exposed group. The mean number of antihypertensive drug refills during the 9 months before the intervention period was higher in the exposed group. The mean distance between the participants' home and the clinic, was longer in the non-exposed group. Differences in mean after-before changes in blood pressure between the two groups were adjusted for age and baseline blood pressure.
Effectiveness results Analyses revealed that the health promotion programme was successful in reducing blood pressure in a sub-group of participants. In individuals with a high family income, a statistically significant mean net reduction of around 8 mmHg in systolic blood pressure was associated with exposure to the programme (p=0.01).
The programme was not successful in decreasing blood pressure in individuals from the low-income strata.
Clinical conclusions The study showed that the pharmacy-based programme was successful in reducing blood pressure in some groups of the population (i.e. those with a higher income).
Measure of benefits used in the economic analysis The measure of benefits used was the monetary benefits derived from the willingness-to-pay (WTP) approach. Participants were asked to state their maximum WTP using the following open-ended question: "Suppose that the pharmacist can improve your quality of life by taking your blood pressure on a regular basis and by advising you on the best way to take your medication. In addition to what you actually pay for your drug insurance coverage, how much would you be willing to pay per month to benefit from this quality of life improvement?".
Direct costs The resource quantities and the costs were reported separately. The direct costs included were those to the pharmacy, the health care system and the patient. These were for antihypertensive drugs, physician visits, hospitalisations, patient travel, pharmacist time, participant (and their relatives or friends) time, and the fixed costs of the intervention and the costs of pharmacist intervention. The latter (pharmacist intervention) comprised the cost of blood readings, verbal interventions and pharmaceutical opinions.
The quantities were extracted from different sources. For example, the RAMQ database (social security in Quebec) was used to obtain the number of physician visits made by each participant and the number of visits at the clinic. Pharmacy files of participants were used to obtain information on the number of pills and refills, the dosage and the name of the antihypertensive medication. Hospitalisation costs were derived from the provincial ministry of health. Travel costs of the participants were derived from participants in the study who were asked their usual means of transportation, and the distance they had to travel. Time spent by the participants going to the pharmacy or to the clinic, as well as their hourly wage, was collected during the interviews to derive time costs. For the 10 participants who did not report their wage, data from a published study was used to attribute a wage. Time spent by pharmacists in refilling antihypertensive medications and administering the programme was derived from the pharmacists included in the study. Estimates for the pharmacists' hourly salary were based on a 1999 annual survey undertaken by the Association of Property Pharmacists of Quebec. The unit costs were derived the same sources used to assess the resource quantities and from participants' questionnaires.
Discounting was not performed. However, the fixed costs were discounted using a 5% discount rate for a 3-year period. The mean costs were reported. The price year was 1998.
Statistical analysis of costs The authors calculated mean treatment costs with their standard deviations. Since the within-group mean cost-differences were skewed, the authors used a 1,000 bootstrap replication of the original data to obtain a p-value. Changes were considered statistically significant when the p-value was equal or less than 0.05.
Indirect Costs The indirect costs (arising from reduced economic productivity) were not included in the analysis.
Sensitivity analysis The authors used the population census data for Quebec province and the percentage of people with hypertension in the province to estimate the number of individuals who might benefit from the programme. The authors performed scenario analysis using two scenarios. One scenario was that of a publicly supported programme, where all individuals would be exposed to the programme and 0% benefit would be made. The other scenario was that of a privately supported programme, where 10% of the hypertensive population would be exposed to the programme and the pharmacists would make a 50% profit on the fixed costs of the programme.
Estimated benefits used in the economic analysis In both groups, the proportion of participants who answered the WTP question before the intervention was at least twice that of the proportion of those who did so after the intervention period. The number of participants answering this question was 31 before the intervention versus 12 after the intervention in the exposed group (n=41), and 38 (before) versus 19 (after) in the non-exposed group (n=59).
In the exposed group, the participants were willing to pay Can$3.29 before the intervention and Can$0.54 after the intervention.
In the non-exposed group, the participants were willing to pay Can$3.00 before the intervention and Can$2.14 after the intervention.
For the entire 9-month period, WTP of exposed participants was Can$4.86.
Cost results Compared with baseline, the mean cost of pharmacist time increased during the intervention period in the exposed group (Can$29.8 before versus Can$62.2 during; p<0.001). The mean direct non-medical costs also increased in this group (Can$37.9 before versus Can$69.7 during; p<0.001). No significant change was observed in the non-exposed group.
On average, the pharmacist intervention costs (i.e. fixed costs, blood pressure readings, verbal interventions, and pharmaceutical opinion) amounted to Can$30.68 per individual exposed to the programme.
Compared with the non-exposed participants, those exposed had a significant decrease in mean direct costs (difference of Can$331.3; p=0.032) and a significant increase in direct non-medical costs (difference of Can$40.7; p<0.001).
The total savings for the exposed participants were $290.6.
Synthesis of costs and benefits The authors determined the cost-benefit of the intervention programme using the benefit (sum of WTP and saved costs) minus both the fixed and pharmacist intervention costs. The authors found the total benefits of the intervention to be Can$295.46 per patient (i.e. WTP per participant for 9 months and the savings achieved by the intervention programme). The authors found the cost-benefits to be -Can$264.54 in the scenario where the programme would be privately supported (scenario 1), and -Can$264.76 in the scenario where the programme would be publicly supported (scenario 2).
The costs of the intervention were Can$30.92 for scenario 1 and Can$30.70 for scenario 2. Hence, the ratio of costs-to-benefits (with either scenario 1 or scenario 2) was 1:9.6 (i.e. the benefits of the programme were about 10-fold higher than the costs).
Authors' conclusions The implementation of a pharmacy-based health promotion programme in hypertension appeared promising, in terms of both costs and benefits, in the quest to improve blood pressure control.
CRD COMMENTARY - Selection of comparators A justification was given for the comparator used. Only 16% of Canadians with hypertension aged between 18 to 74 years are treated and have their blood pressure controlled. You should decide if this is a widely used health technology in your own setting.
Validity of estimate of measure of effectiveness The basis of the analysis was a quasi-experimental cohort study in nine pharmacies in Quebec City area. The authors stated that this study design minimises the risk of contamination of the programme in the non-exposed group. Although the assignment of pharmacies and patients to exposure or usual care was not randomised, the authors reported that this should not be viewed as a limitation or threat to the validity of the study, as the motivation of those people who performed the intervention was critical to the appropriate implementation of the programme. The authors reported that their study population seemed different from the hypertensive population of the province of Quebec, as fewer participants in the study had their blood pressure controlled at baseline than the rest of the population. In addition, characteristics such as age and gender were different. The patient groups were shown to be generally comparable in terms of baseline characteristics. The exceptions were the proportion using a car, the number of drug refills, and the mean distance between the participants' home and the clinic. Appropriate statistical analyses were undertaken. These tested for statistically significant differences between mean differences between both groups, and between mean differences before and after the intervention within the groups. Further, interviewers were blinded to the group to which the patient belonged.
Validity of estimate of measure of benefit The estimation of benefits was obtained directly from participants using the WTP approach. The choice of using monetary benefits was justified as the authors undertook a cost-benefit analysis.
Validity of estimate of costs Although the authors reported that a societal perspective was adopted, the productivity costs were not included in the analysis, and no explanation was given for their exclusion. The authors reported that training costs were not included in the analysis, hence slightly underestimating the costs of the pharmacist programme. The costs and the quantities were reported separately, which will enhance the generalisability of the authors' results. Resource use was obtained from the effectiveness study and through questionnaires. The authors reported that the use of databases and pharmacy files enabled them to avoid the potential bias related to the participants' memory. The costs were obtained from a variety of sources such as the authors' setting and published sources. Differences in costs between the two groups were appropriately tested for statistical significance using bootstrap replications. Discounting was not relevant, as the treatment costs were incurred during 9 months, and was not performed. As the fixed costs were incurred over a longer time period than the other costs, which were incurred during 9 months, these costs were appropriately discounted. The price year was reported, which will aid any possible inflation exercises.
Other issues The authors reported that no study had assessed the impact of pharmacist interventions on cost and benefit using WTP as part of benefit. However, four studies have assessed the impact of pharmacist intervention on cost. These all showed that, no matter what the medical condition, pharmacist intervention resulted in cost-savings, as was shown in the present study. The issue of generalisability to other settings was partly addressed by the authors comparing the costs of the programme under two different provision scenarios (a publicly funded programme and a privately funded programme). The authors do not appear to have presented their results selectively and their conclusions reflected the scope of the analysis.
The authors reported further limitations to their study. First, the time cost for the pharmacists who delivered the intervention might have been overestimated. Second, the training costs were not included in the analysis, therefore slightly underestimating the costs of the programme. Finally, the study population was different to the population in Quebec province, although the authors reported that they would expect smaller mean treatment costs, thus resulting in a greater benefit from the programme, if the programme was implemented at the provincial level.
Implications of the study The authors reported that the findings from this study warrant more research to assess the long-term clinical and economic impact of the pharmacy-based programme on a larger population and for a longer period of time.
Source of funding Funded in part by the Fonds d'Enseignement et de Recherche, Faculte de Pharmacie, Universite Laval, Quebec, Canada.
Bibliographic details Cote I, Gregoire J-P, Moisan J, Chabot I, Lacroix G. A pharmacy-based health promotion programme in hypertension: cost-benefit analysis. PharmacoEconomics 2003; 21(6): 415-428 Other publications of related interest Chabot I. Impact d'un programme d'intervention sur la tension arterielle d'hypertendus traites pharmacologiquement en pharmacie communautaire. Une etude exploratoire. Quebec: Universite Laval; 2001.
Munroe WP, Kunz K, Dalmady-Israel C, et al. Economic evaluation of pharmacist involvement in disease management in a community pharmacy setting. Clinical Therapeutics 1997;19:113-23.
Jameson J, VanNoord G, Vanderwoud K. The impact of pharmacotherapy consultation on the cost and outcome of medical therapy. Journal of Family Practice 1995;41:469-72.
Kroger E, Moisan J, Gregoire JP. Billing for cognitive services: understanding Quebec pharmacists' behaviour. Annals of Pharmacotherapy 2000;34:309-16.
Indexing Status Subject indexing assigned by NLM MeSH Age Factors; Aged; Cost-Benefit Analysis; Female; Health Promotion /economics; Humans; Hypertension /economics /therapy; Male; Middle Aged; Pharmacies /economics; Pharmacists; Quebec; Surveys and Questionnaires AccessionNumber 22003008125 Date bibliographic record published 31/05/2005 Date abstract record published 31/05/2005 |
|
|
|