Pharmacy computer prescription databases: methodologic issues of access and confidentiality

Public opinions about participating in health research

OBJECTIVES

Privacy legislation has limited options for recruiting subjects to health studies. Policy changes are motivated by assumptions about public attitudes towards participation, yet surveys of attitudes have rarely been done. We investigated public willingness to participate in health research and how willingness was affected by various factors.

METHODS

A survey of adults randomly selected from the telephone directory was conducted in British Columbia, Canada. Mailed self-administered questionnaires asked about willingness to participate in health research and the influence on willingness of the method of subject selection, the organization making the contact, and other factors.

RESULTS

There were 1,477 respondents (58% of eligible); 85% were willing to participate in health research at least sometimes. The organization making the contact influenced comfort about participation: 10% of respondents felt uncomfortable if contacted by a university, 12% if by a hospital, 26% if by government, and 55% if by private research firms. Factors most positively influencing choice to participate were future health benefits to society (87%) and oneself (87%), and receiving a copy of the study results (81%).

CONCLUSIONS

Participation in health research appears to be viewed favourably by members of the public, and participation may be highest when university or hospital-based researchers are able to contact subjects directly using information from government databases.

Clinical and humanistic outcomes of a lipid management program in the community pharmacy setting

OBJECTIVES

To identify patients at risk for coronary artery disease (CAD) through a search of a community pharmacy's prescription database, to screen and identify patients with elevated cholesterol and at risk for CAD, to enroll patients in a pharmacist-directed lipid management program, and to evaluate selected clinical and humanistic outcomes.

DESIGN

Randomized, pretest-posttest control groups.

SETTING

Independent community pharmacy in a suburban metropolitan area.

PATIENTS

51 patients who were not at National Cholesterol Education Program low-density lipoprotein cholesterol (LDL-C) or defined triglyceride goals and who met inclusion criteria.

INTERVENTION

Pharmacist-directed lipid management program.

MAIN OUTCOME MEASURES

Clinical outcome measures included total cholesterol, LDL-C, high-density lipoprotein cholesterol (HDL-C), and triglyceride levels; achievement of LDL-C goal; and risk factor prediction scores. Humanistic outcome measures included patient satisfaction with pharmaceutical care and patient knowledge of hyperlipidemia.

RESULTS

LDL-C was decreased in the pharmacist intervention group (n = 25), compared with an increase in the control group at study end. HDL-C levels increased and triglyceride levels decreased in both groups. Of treatment group patients, 32% achieved their cholesterol goals, compared with 15% of control group patients. Risk factor prediction scores improved in the treatment group and worsened in the control group. The treatment group's hyperlipidemia knowledge scores improved significantly from pretest to posttest. Both treatment and control group patient satisfaction scores for the pharmacist investigator were favorable at study end.

CONCLUSION

Both treatment and control patients benefited from participating in this study. Patients enrolled in the lipid management program made greater improvements in their knowledge of hyperlipidemia, risk factor scores, and cholesterol levels.

A community-pharmacy-based callback program for antibiotic therapy

OBJECTIVE

To develop a telephone callback program to reinforce basic patient counseling given at the time of dispensing.

DESIGN

At the time each prescription was filled, the pharmacist or pharmacy student recorded the patient's name, telephone number, date of birth, diagnosis, drug regimen, prescriber, relevant concurrent medications, initial comments, and date. Follow-up phone calls were then made when appropriate to assess patient progress in therapy.

RESULTS

Telephone assessment of 521 patients revealed that while 51.1% reported their condition had improved, 15.1% had worsened. Of the 79 therapeutic failures, 47 patients repeated the initial therapy or received a different antibiotic without another physician visit, 32 were referred back to their physicians, and 5 each resulted from noncompliance or adverse effects.

CONCLUSIONS

A callback program is an effective and inexpensive mechanism for assessing and improving drug therapy outcomes.

Use of pharmacy computer prescription database to access hypertensive patients for mailed survey research

OBJECTIVE

To evaluate the use of a pharmacy computer prescription database (PCPD) to identify and sample hypertensive patients outside of their treatment setting for mailed, self-administered survey research comparing the presence and lack of participation incentives and mailed follow-up reminders.

SETTING

Two independent, privately owned, free-standing community pharmacies.

PARTICIPANTS

A total of 735 individuals identified from a PCPD search who had been prescribed at least 1 of 130 PCPD medications potentially used in hypertensive therapy.

METHODS

PCPD was searched by the pharmacist/owner per protocol; participants were sent an individually addressed cover letter on pharmacy letterhead signed by the pharmacist that requested voluntary anonymous completion of an enclosed self-administered, quality-of-life opinion survey; the first search used no incentive or follow-up; the second search used an incentive and mailed a follow-up reminder. Research protocol followed published ethics guidelines.

RESULTS

There was a mean 84% return with incentive and follow-up strategies compared with a mean 25% return without strategies (p < 0.01) for all drug groups between searches; no statistical difference in response was shown between the same drug groups (alpha 1-blockers, calcium-channel blockers, and centrally acting alpha 2-agonists) within searches.

CONCLUSIONS

Acceptable response rates (74-93% return) can be obtained with traditional follow-up mailed incentives, prescription issuance within 6 months, incentives to both hypertensive and nonhypertensive responders, and sampling strategies within potential drug groups. Methodology can be adapted to other populations by alternative drug sampling strategies.