Rapid responses are electronic letters to the editor. They enable our users to debate issues raised in articles published on thebmj.com. Although a selection of rapid responses will be included online and in print as readers' letters, their first appearance online means that they are published articles. If you need the url (web address) of an individual response, perhaps for citation purposes, simply click on the response headline and copy the url from the browser window. Letters are indexed in PubMed.
The Holland et al. [BMJ, doi:10.1136/bmj.38338.674583.AE (published
24 January 2005)] findings that pharmacist medication reviews for home-
dwelling patients are associated with a 30% increase in emergency
hospitalizations and a greater reduction in global health score (as
measured on a Visual Analog Scale) is paradoxical, counterintuitive and
inconsistent with the U.S. literature on pharmacist management of
medication therapy.[1-3] The authors attribute these results to either 1)
an increase in health care-seeking behavior by intervention patients who
become better-sensitized to their health conditions, 2) an increase in
hospitalizations among frail intervention patients who otherwise would
have died, 3) an increase in iatrogenic medication adherence among
intervention patients, and/or 4) an increase in intervention patient
confusion and anxiety about health care services.
While all of these explanations imply strong pharmacist intervention
effects (good or bad), it would be valuable to have more details about the
patient sample and their reasons for hospital readmission. 11 of the
excess hospitalizations in the pharmacist intervention group (20%) were
accounted for by just two of the 429 intervention patients and 15 excess
hospitalizations (27%) were accounted for by 5 other intervention
patients. The patients were not stratified by baseline diagnosis with the
unfortunate result that nearly twice as many cancer (Odd Ratio 2.1,
p<_.05 one-tailed="one-tailed" and="and" neurological="neurological" stroke="stroke" senility="senility" or="or" dementia="dementia" odd="odd" ratio="ratio" _1.6="_1.6" p.05="p.05" patients="patients" were="were" assigned="assigned" to="to" the="the" intervention="intervention" group="group" as="as" control="control" group.="group." none="none" of="of" other="other" baseline="baseline" diagnosis="diagnosis" differences="differences" appear="appear" significant="significant" across="across" treatment="treatment" groups="groups" but="but" because="because" their="their" high="high" relative="relative" hospitalization="hospitalization" propensity="propensity" it="it" would="would" be="be" useful="useful" see="see" if="if" main="main" findings="findings" still="still" obtain="obtain" when="when" cancer="cancer" are="are" excluded="excluded" from="from" analysis.="analysis." p="p"/> This sample is atypical in that all patients were quite elderly (mean
age 85.5) averaged 5.9 drugs daily, and were recruited from an initial
inpatient admission. Patients admitted to the hospital initially because
of iatrogenic medication prescribing, or other issues of relatively poor
health care quality (e.g. inappropriate medical diagnosis or treatment)
would be more likely to be recruited into this study than patients
recruited from the community or from ambulatory settings. It is not clear
that the pharmacist interventions utilized in this study were specifically
designed to address and correct such health care quality issues.
The study certainly raises a very valuable cautionary note that
medication management interventions, particularly in the frail elderly,
and particularly in patients averaging more than five daily medications
need to be designed and planned very carefully to ensure that the
intervention benefits outweigh the risks. But because of these study
design concerns it is not clear how relevant these findings are to the
broader issue of assessing the value of medication therapy management
services.
Sincerely,
Joel W. Hay, PhD
Associate Prof.
Dept. of Pharmaceutical Economics and Policy,
University of Southern California,
School of Pharmacy
1. Etemad L, Hay J. Cost Effectiveness Analysis of Pharmaceutical
Care in a Medicare Drug Benefit Program. Value in Health. 2003, 6(4):425-
435.
2. Yuan Y, Hay JW, McCombs JS. Mortality and hospitalization impacts
of pharmacist consultation in ambulatory care. Am J Manag Care
2003;9:101–12.
3. McCombs JS, Cody M, Parker JP, Johnson KA, Besinque K, Borok G,
Ershoff D, Groshen S, Hay J, Nichol MB, Nye MT. The Kaiser Permanente/USC
Patient Consultation Study: Change in Use and Cost of Health Services. Am
J Health-Syst Pharm 1998. 55(2):2485-99.
Competing interests:
None declared
Competing interests:
No competing interests
28 January 2005
Joel W Hay
Assoc. Professor
University of Southern California School of Pharmacy, Los Angeles CA 90089 USA
Pharmacist Medication Review Study Design Concerns
To The Editor:
The Holland et al. [BMJ, doi:10.1136/bmj.38338.674583.AE (published
24 January 2005)] findings that pharmacist medication reviews for home-
dwelling patients are associated with a 30% increase in emergency
hospitalizations and a greater reduction in global health score (as
measured on a Visual Analog Scale) is paradoxical, counterintuitive and
inconsistent with the U.S. literature on pharmacist management of
medication therapy.[1-3] The authors attribute these results to either 1)
an increase in health care-seeking behavior by intervention patients who
become better-sensitized to their health conditions, 2) an increase in
hospitalizations among frail intervention patients who otherwise would
have died, 3) an increase in iatrogenic medication adherence among
intervention patients, and/or 4) an increase in intervention patient
confusion and anxiety about health care services.
While all of these explanations imply strong pharmacist intervention
effects (good or bad), it would be valuable to have more details about the
patient sample and their reasons for hospital readmission. 11 of the
excess hospitalizations in the pharmacist intervention group (20%) were
accounted for by just two of the 429 intervention patients and 15 excess
hospitalizations (27%) were accounted for by 5 other intervention
patients. The patients were not stratified by baseline diagnosis with the
unfortunate result that nearly twice as many cancer (Odd Ratio 2.1,
p<_.05 one-tailed="one-tailed" and="and" neurological="neurological" stroke="stroke" senility="senility" or="or" dementia="dementia" odd="odd" ratio="ratio" _1.6="_1.6" p.05="p.05" patients="patients" were="were" assigned="assigned" to="to" the="the" intervention="intervention" group="group" as="as" control="control" group.="group." none="none" of="of" other="other" baseline="baseline" diagnosis="diagnosis" differences="differences" appear="appear" significant="significant" across="across" treatment="treatment" groups="groups" but="but" because="because" their="their" high="high" relative="relative" hospitalization="hospitalization" propensity="propensity" it="it" would="would" be="be" useful="useful" see="see" if="if" main="main" findings="findings" still="still" obtain="obtain" when="when" cancer="cancer" are="are" excluded="excluded" from="from" analysis.="analysis." p="p"/> This sample is atypical in that all patients were quite elderly (mean
age 85.5) averaged 5.9 drugs daily, and were recruited from an initial
inpatient admission. Patients admitted to the hospital initially because
of iatrogenic medication prescribing, or other issues of relatively poor
health care quality (e.g. inappropriate medical diagnosis or treatment)
would be more likely to be recruited into this study than patients
recruited from the community or from ambulatory settings. It is not clear
that the pharmacist interventions utilized in this study were specifically
designed to address and correct such health care quality issues.
The study certainly raises a very valuable cautionary note that
medication management interventions, particularly in the frail elderly,
and particularly in patients averaging more than five daily medications
need to be designed and planned very carefully to ensure that the
intervention benefits outweigh the risks. But because of these study
design concerns it is not clear how relevant these findings are to the
broader issue of assessing the value of medication therapy management
services.
Sincerely,
Joel W. Hay, PhD
Associate Prof.
Dept. of Pharmaceutical Economics and Policy,
University of Southern California,
School of Pharmacy
1. Etemad L, Hay J. Cost Effectiveness Analysis of Pharmaceutical
Care in a Medicare Drug Benefit Program. Value in Health. 2003, 6(4):425-
435.
2. Yuan Y, Hay JW, McCombs JS. Mortality and hospitalization impacts
of pharmacist consultation in ambulatory care. Am J Manag Care
2003;9:101–12.
3. McCombs JS, Cody M, Parker JP, Johnson KA, Besinque K, Borok G,
Ershoff D, Groshen S, Hay J, Nichol MB, Nye MT. The Kaiser Permanente/USC
Patient Consultation Study: Change in Use and Cost of Health Services. Am
J Health-Syst Pharm 1998. 55(2):2485-99.
Competing interests:
None declared
Competing interests: No competing interests