Which Medicines Do Our Patients Want From Us?

As health maintenance organizations (HMOs) and physicians increasingly become medical partners, the HMO formularies and prescribing recommendations will be more evident in physicians’ practices. Upper-respiratory infections and low back pain are so common and often so difficult to treat that they are prime targets for HMO intervention.

A recent study showed that HMO physicians, perform less diagnostic testing, but may prescribe more antibiotics for viral illnesses than other physicians.¹ Often the physician believes the patient expects antibiotics when that is not the case.^2-5 In turn, patients may feel they need treatment for self-limited conditions because they have received them in the past.⁶

Physicians may believe that patients will be dissatisfied when expectations are not met. Dissatisfied patients are less likely to comply with physicians’ treatment recommendations.⁷ Patients may continue to call or visit the physician until expectations are met.⁸ Sometimes physicians think that patients will go elsewhere to get treatment. We asked patients for their opinions on the need for prescriptive medications and compared prepaid and fee-for-service (FFS) patients.

Methods

A total of 244 consecutive patients or parents of patients in an urban family medicine clinic completed survey forms in October 1997. This was an undifferentiated patient population seeking care for any reason. No attempt was made to screen for the reason for the visit. Parents completed the questionnaire for any child in the consecutive grouping. The questionnaire addressed perceptions of appropriateness of the frequency of medication prescription in general, reassurance or prescriptions as treatments related to 3 specific problems (virus, muscle straining, diarrhea), and preferred timing of treatment modification. No one refused to complete the questionnaire.

Men made up 37% of the sample (n=91) and women 63% (n=153). Patients were mostly white (n=191, 78%), with a substantial African American minority group (n=50, 21%), and 1% were of other ethnicities. Eighteen percent of the sample were single (n=44); 66%, married (n=160); 7%, widowed (n=16); and 9%, divorced or separated (n=23). Most of the patients were aged 30 to 59 years, with 17% 60 years or older and 20% 29 years or younger.

Sixty-seven percent of the sample were HMO members and 30% were fee-for-service patients. Most of the HMO members were charged a flat fee for medications (63%). White patients were more likely to be enrolled in an HMO than African Americans (72% vs 59%, respectively; c2= 2.937; P <087).

Analyses were largely descriptive with chi-squares conducted for group comparisons of categorical variables. Respondents were excluded from analyses if any item was missing. We conducted 2-way analysis of variance on continuous treatment change questions to compare race and HMO status.

Results

Medication Questions

Most respondents indicated that they did not believe physicians prescribed too many medications (75%). Eighty-four percent of the patients would want an expensive medicine if the physician recommended it. In a comparison of patients who had a set fee for medicine with those who did not, 88% with the set fee said they would want the more expensive medications as opposed to 77% in the group without a set fee (c2= 3.39, P <.066). African Americans were less likely to want the more costly medicine than were whites (79.6% vs 92%, respectively; c2= 6.471; P <.011).

Reassurance and Prescription Questions

We asked several questions regarding personal preferences about receiving prescriptions, reassurance, or antibiotics for virus, muscle strains, and diarrhea. Reassurance and over-the-counter treatments were seen as sufficient care for viruses (84%), muscle strain (72%), and diarrhea (56%). In contrast, when asked if they would want an antibiotic or pain medication “just to be sure,” 43% said yes in the case of a virus, 37% for a muscle strain, and 92% for diarrhea.

For viruses, we found a significant difference in preferences (P <.001) by race, with white patients more likely to accept reassurance as sufficient care (76%) than African Americans (51%). African Americans were also less likely than whites to see reassurance as sufficient for muscle strain (P <001) and diarrhea (P <.06). Table 1 shows the percentages of patients accepting reassurance. When asked about prescriptions, “just to be sure,” race and HMO membership influenced the outcome, with more African Americans than whites likely to want an antibiotic for a virus (P <001) as were more FFS members than HMO members (P <.001). For a muscle strain, FFS members and African Americans were again more likely to want painkillers “just to be sure.” For diarrhea, FFS respondents were more likely than HMO respondents to want an antibiotic (P <.03), and there was a similar trend for African Americans to prefer the medication (P <.06). Table 2 shows the percentage of patients wanting medication “just to be sure.”

Eighty-four percent of the sample said that reassurance and cold medicine would be enough care for a virus that antibiotics would not help. A significant difference was found by race, with white patients more likely to accept reassurance (76%) than African Americans (51%, X²=11.176, P <.001).

Forty-three percent of the sample would want an antibiotic “just to be sure” if they had a virus (even though 84% said reassurance was sufficient). Sex, race, and HMO membership all influenced this finding with women, African Americans, and patients not in an HMO more likely to want the antibiotic (56% women vs 39% of men, X²=6.020, P <.014; 71% of African-Americans vs 39% of whites, c2= 15.348, P <000; 63% of patients not in an HMO vs 38% of HMO members, X²=11.712, P <.001).

Similar questions were asked in reference to muscle strain: would reassurance be enough? Again, African Americans were less likely to see reassurance as sufficient (55% vs 81%, X²=14.003, P <.000). Seventy-two percent of those responding felt this would be sufficient care. However, 37% stated that they would like muscle relaxers or pain medication just to be sure. Patients who were in an HMO and African Americans were more likely to prefer the medication than those not in an HMO (55% HMO vs 36% non-HMO, X²=6.014, P<.014; 63% African Americans vs 37% whites, X²=9.59, P<.002).

Finally, the same questions were asked about diarrhea, which was most likely caused by a virus. Reassurance was again less apt to be perceived as sufficient by African Americans (50% vs 65%, X²= 3.442, P<.064), and they were more likely to want an antibiotic “just to be sure” (37% vs 23%, X²=3.759, P<.053). The patients not in an HMO were also more likely to want an antibiotic just to be sure (35% vs 21%, X²=4.669, P<.031).

Treatment Change Questions

Two questions were asked about the timing of requests for different treatments. The first had to do with the timing of a change in medicine (assuming the patient was not getting sicker) and the second with the number of medication changes that would be acceptable before requesting a referral to a specialist. Overall, 14% of the respondents would call for new medicine within 24 hours; 22% would call within 48 hours; 31% would call within 72 hours; and 33% would wait longer. In terms of requesting a specialist, 8% of the respondents would accept one medicine change before calling a specialist; 43% would accept 2 changes; 36% would accept 3 changes; and 13% would accept more changes. African Americans and FFS patients would wait for a shorter amount of time before calling for a medication change (F=14.66, P <017; F=6.34, P <013, respectively). No differences were observed by sex, race, or HMO membership status in number of medication changes before going to a specialist.

Discussion

We addressed patient preferences concerning 3 main topics: (1) influence of cost of medicine, (2) use of reassurance as a valid treatment, and (3) timing of treatment changes. Results indicate that patients want expensive medicine if there is a reason to believe it will work better. This finding is particularly true of HMO enrollees who have a set charge for medicine, suggesting that such plans and provisions may change patient preferences in a more costly manner or that patients self-select for membership. The acceptance of reassurance as a valid treatment by many patients belies the stereotype of patients always wanting prescriptions. However, variation in the wording of that question (the switch to whether they would want medication “just to be sure”) significantly reduced the percentage of those who simply accepted over-the-counter treatments and reassurance as adequate treatments. Substantial variation was observed by race, with African Americans consistently not as accepting of reassurance and more desirous of “just to be sure” medication. Although the small sample size precludes the drawing of conclusions, this may be suggestive of cultural differences in health-seeking behavior and also reduced exposure to managed care plans. One possibility is that African Americans may have a higher threshold for consulting a physician and may have tried symptomatic treatment for a longer period of time than whites. Because FFS patients were more likely to be African American, observed differences by race may be because of the covariation of race with insurer type. Future studies should explore these possible explanations.

We also found that patients are willing to accept multiple changes in medicines before seeking specialist care and to wait several days before calling physicians about medication ineffectiveness. African Americans and FFS members were less willing to wait. Perhaps there is a self-selection factor to enrollment in HMOs, or perhaps HMO members change their expectations after enrollment. In any case, these data show that understanding the role of patient preferences and expectations about the care they receive may have some potential for cost saving through the use of reassurance and corresponding patient satisfaction with care that does not always deliver a prescription.

Acknowledgements

The authors wish to thank Pam Stovall and Tracey Barton for their typing of the manuscript.

As health maintenance organizations (HMOs) and physicians increasingly become medical partners, the HMO formularies and prescribing recommendations will be more evident in physicians’ practices. Upper-respiratory infections and low back pain are so common and often so difficult to treat that they are prime targets for HMO intervention.

A recent study showed that HMO physicians, perform less diagnostic testing, but may prescribe more antibiotics for viral illnesses than other physicians.¹ Often the physician believes the patient expects antibiotics when that is not the case.^2-5 In turn, patients may feel they need treatment for self-limited conditions because they have received them in the past.⁶

Physicians may believe that patients will be dissatisfied when expectations are not met. Dissatisfied patients are less likely to comply with physicians’ treatment recommendations.⁷ Patients may continue to call or visit the physician until expectations are met.⁸ Sometimes physicians think that patients will go elsewhere to get treatment. We asked patients for their opinions on the need for prescriptive medications and compared prepaid and fee-for-service (FFS) patients.

Methods

A total of 244 consecutive patients or parents of patients in an urban family medicine clinic completed survey forms in October 1997. This was an undifferentiated patient population seeking care for any reason. No attempt was made to screen for the reason for the visit. Parents completed the questionnaire for any child in the consecutive grouping. The questionnaire addressed perceptions of appropriateness of the frequency of medication prescription in general, reassurance or prescriptions as treatments related to 3 specific problems (virus, muscle straining, diarrhea), and preferred timing of treatment modification. No one refused to complete the questionnaire.

Men made up 37% of the sample (n=91) and women 63% (n=153). Patients were mostly white (n=191, 78%), with a substantial African American minority group (n=50, 21%), and 1% were of other ethnicities. Eighteen percent of the sample were single (n=44); 66%, married (n=160); 7%, widowed (n=16); and 9%, divorced or separated (n=23). Most of the patients were aged 30 to 59 years, with 17% 60 years or older and 20% 29 years or younger.

Sixty-seven percent of the sample were HMO members and 30% were fee-for-service patients. Most of the HMO members were charged a flat fee for medications (63%). White patients were more likely to be enrolled in an HMO than African Americans (72% vs 59%, respectively; c2= 2.937; P <087).

Analyses were largely descriptive with chi-squares conducted for group comparisons of categorical variables. Respondents were excluded from analyses if any item was missing. We conducted 2-way analysis of variance on continuous treatment change questions to compare race and HMO status.

Results

Medication Questions

Most respondents indicated that they did not believe physicians prescribed too many medications (75%). Eighty-four percent of the patients would want an expensive medicine if the physician recommended it. In a comparison of patients who had a set fee for medicine with those who did not, 88% with the set fee said they would want the more expensive medications as opposed to 77% in the group without a set fee (c2= 3.39, P <.066). African Americans were less likely to want the more costly medicine than were whites (79.6% vs 92%, respectively; c2= 6.471; P <.011).

Reassurance and Prescription Questions

We asked several questions regarding personal preferences about receiving prescriptions, reassurance, or antibiotics for virus, muscle strains, and diarrhea. Reassurance and over-the-counter treatments were seen as sufficient care for viruses (84%), muscle strain (72%), and diarrhea (56%). In contrast, when asked if they would want an antibiotic or pain medication “just to be sure,” 43% said yes in the case of a virus, 37% for a muscle strain, and 92% for diarrhea.

For viruses, we found a significant difference in preferences (P <.001) by race, with white patients more likely to accept reassurance as sufficient care (76%) than African Americans (51%). African Americans were also less likely than whites to see reassurance as sufficient for muscle strain (P <001) and diarrhea (P <.06). Table 1 shows the percentages of patients accepting reassurance. When asked about prescriptions, “just to be sure,” race and HMO membership influenced the outcome, with more African Americans than whites likely to want an antibiotic for a virus (P <001) as were more FFS members than HMO members (P <.001). For a muscle strain, FFS members and African Americans were again more likely to want painkillers “just to be sure.” For diarrhea, FFS respondents were more likely than HMO respondents to want an antibiotic (P <.03), and there was a similar trend for African Americans to prefer the medication (P <.06). Table 2 shows the percentage of patients wanting medication “just to be sure.”

Eighty-four percent of the sample said that reassurance and cold medicine would be enough care for a virus that antibiotics would not help. A significant difference was found by race, with white patients more likely to accept reassurance (76%) than African Americans (51%, X²=11.176, P <.001).

Forty-three percent of the sample would want an antibiotic “just to be sure” if they had a virus (even though 84% said reassurance was sufficient). Sex, race, and HMO membership all influenced this finding with women, African Americans, and patients not in an HMO more likely to want the antibiotic (56% women vs 39% of men, X²=6.020, P <.014; 71% of African-Americans vs 39% of whites, c2= 15.348, P <000; 63% of patients not in an HMO vs 38% of HMO members, X²=11.712, P <.001).

Similar questions were asked in reference to muscle strain: would reassurance be enough? Again, African Americans were less likely to see reassurance as sufficient (55% vs 81%, X²=14.003, P <.000). Seventy-two percent of those responding felt this would be sufficient care. However, 37% stated that they would like muscle relaxers or pain medication just to be sure. Patients who were in an HMO and African Americans were more likely to prefer the medication than those not in an HMO (55% HMO vs 36% non-HMO, X²=6.014, P<.014; 63% African Americans vs 37% whites, X²=9.59, P<.002).

Finally, the same questions were asked about diarrhea, which was most likely caused by a virus. Reassurance was again less apt to be perceived as sufficient by African Americans (50% vs 65%, X²= 3.442, P<.064), and they were more likely to want an antibiotic “just to be sure” (37% vs 23%, X²=3.759, P<.053). The patients not in an HMO were also more likely to want an antibiotic just to be sure (35% vs 21%, X²=4.669, P<.031).

Treatment Change Questions

Two questions were asked about the timing of requests for different treatments. The first had to do with the timing of a change in medicine (assuming the patient was not getting sicker) and the second with the number of medication changes that would be acceptable before requesting a referral to a specialist. Overall, 14% of the respondents would call for new medicine within 24 hours; 22% would call within 48 hours; 31% would call within 72 hours; and 33% would wait longer. In terms of requesting a specialist, 8% of the respondents would accept one medicine change before calling a specialist; 43% would accept 2 changes; 36% would accept 3 changes; and 13% would accept more changes. African Americans and FFS patients would wait for a shorter amount of time before calling for a medication change (F=14.66, P <017; F=6.34, P <013, respectively). No differences were observed by sex, race, or HMO membership status in number of medication changes before going to a specialist.

Discussion

We addressed patient preferences concerning 3 main topics: (1) influence of cost of medicine, (2) use of reassurance as a valid treatment, and (3) timing of treatment changes. Results indicate that patients want expensive medicine if there is a reason to believe it will work better. This finding is particularly true of HMO enrollees who have a set charge for medicine, suggesting that such plans and provisions may change patient preferences in a more costly manner or that patients self-select for membership. The acceptance of reassurance as a valid treatment by many patients belies the stereotype of patients always wanting prescriptions. However, variation in the wording of that question (the switch to whether they would want medication “just to be sure”) significantly reduced the percentage of those who simply accepted over-the-counter treatments and reassurance as adequate treatments. Substantial variation was observed by race, with African Americans consistently not as accepting of reassurance and more desirous of “just to be sure” medication. Although the small sample size precludes the drawing of conclusions, this may be suggestive of cultural differences in health-seeking behavior and also reduced exposure to managed care plans. One possibility is that African Americans may have a higher threshold for consulting a physician and may have tried symptomatic treatment for a longer period of time than whites. Because FFS patients were more likely to be African American, observed differences by race may be because of the covariation of race with insurer type. Future studies should explore these possible explanations.

We also found that patients are willing to accept multiple changes in medicines before seeking specialist care and to wait several days before calling physicians about medication ineffectiveness. African Americans and FFS members were less willing to wait. Perhaps there is a self-selection factor to enrollment in HMOs, or perhaps HMO members change their expectations after enrollment. In any case, these data show that understanding the role of patient preferences and expectations about the care they receive may have some potential for cost saving through the use of reassurance and corresponding patient satisfaction with care that does not always deliver a prescription.

Acknowledgements

The authors wish to thank Pam Stovall and Tracey Barton for their typing of the manuscript.

As health maintenance organizations (HMOs) and physicians increasingly become medical partners, the HMO formularies and prescribing recommendations will be more evident in physicians’ practices. Upper-respiratory infections and low back pain are so common and often so difficult to treat that they are prime targets for HMO intervention.

A recent study showed that HMO physicians, perform less diagnostic testing, but may prescribe more antibiotics for viral illnesses than other physicians.¹ Often the physician believes the patient expects antibiotics when that is not the case.^2-5 In turn, patients may feel they need treatment for self-limited conditions because they have received them in the past.⁶

Physicians may believe that patients will be dissatisfied when expectations are not met. Dissatisfied patients are less likely to comply with physicians’ treatment recommendations.⁷ Patients may continue to call or visit the physician until expectations are met.⁸ Sometimes physicians think that patients will go elsewhere to get treatment. We asked patients for their opinions on the need for prescriptive medications and compared prepaid and fee-for-service (FFS) patients.

Methods

A total of 244 consecutive patients or parents of patients in an urban family medicine clinic completed survey forms in October 1997. This was an undifferentiated patient population seeking care for any reason. No attempt was made to screen for the reason for the visit. Parents completed the questionnaire for any child in the consecutive grouping. The questionnaire addressed perceptions of appropriateness of the frequency of medication prescription in general, reassurance or prescriptions as treatments related to 3 specific problems (virus, muscle straining, diarrhea), and preferred timing of treatment modification. No one refused to complete the questionnaire.

Men made up 37% of the sample (n=91) and women 63% (n=153). Patients were mostly white (n=191, 78%), with a substantial African American minority group (n=50, 21%), and 1% were of other ethnicities. Eighteen percent of the sample were single (n=44); 66%, married (n=160); 7%, widowed (n=16); and 9%, divorced or separated (n=23). Most of the patients were aged 30 to 59 years, with 17% 60 years or older and 20% 29 years or younger.

Sixty-seven percent of the sample were HMO members and 30% were fee-for-service patients. Most of the HMO members were charged a flat fee for medications (63%). White patients were more likely to be enrolled in an HMO than African Americans (72% vs 59%, respectively; c2= 2.937; P <087).

Analyses were largely descriptive with chi-squares conducted for group comparisons of categorical variables. Respondents were excluded from analyses if any item was missing. We conducted 2-way analysis of variance on continuous treatment change questions to compare race and HMO status.

Results

Medication Questions

Most respondents indicated that they did not believe physicians prescribed too many medications (75%). Eighty-four percent of the patients would want an expensive medicine if the physician recommended it. In a comparison of patients who had a set fee for medicine with those who did not, 88% with the set fee said they would want the more expensive medications as opposed to 77% in the group without a set fee (c2= 3.39, P <.066). African Americans were less likely to want the more costly medicine than were whites (79.6% vs 92%, respectively; c2= 6.471; P <.011).

Reassurance and Prescription Questions

We asked several questions regarding personal preferences about receiving prescriptions, reassurance, or antibiotics for virus, muscle strains, and diarrhea. Reassurance and over-the-counter treatments were seen as sufficient care for viruses (84%), muscle strain (72%), and diarrhea (56%). In contrast, when asked if they would want an antibiotic or pain medication “just to be sure,” 43% said yes in the case of a virus, 37% for a muscle strain, and 92% for diarrhea.

For viruses, we found a significant difference in preferences (P <.001) by race, with white patients more likely to accept reassurance as sufficient care (76%) than African Americans (51%). African Americans were also less likely than whites to see reassurance as sufficient for muscle strain (P <001) and diarrhea (P <.06). Table 1 shows the percentages of patients accepting reassurance. When asked about prescriptions, “just to be sure,” race and HMO membership influenced the outcome, with more African Americans than whites likely to want an antibiotic for a virus (P <001) as were more FFS members than HMO members (P <.001). For a muscle strain, FFS members and African Americans were again more likely to want painkillers “just to be sure.” For diarrhea, FFS respondents were more likely than HMO respondents to want an antibiotic (P <.03), and there was a similar trend for African Americans to prefer the medication (P <.06). Table 2 shows the percentage of patients wanting medication “just to be sure.”

Eighty-four percent of the sample said that reassurance and cold medicine would be enough care for a virus that antibiotics would not help. A significant difference was found by race, with white patients more likely to accept reassurance (76%) than African Americans (51%, X²=11.176, P <.001).

Forty-three percent of the sample would want an antibiotic “just to be sure” if they had a virus (even though 84% said reassurance was sufficient). Sex, race, and HMO membership all influenced this finding with women, African Americans, and patients not in an HMO more likely to want the antibiotic (56% women vs 39% of men, X²=6.020, P <.014; 71% of African-Americans vs 39% of whites, c2= 15.348, P <000; 63% of patients not in an HMO vs 38% of HMO members, X²=11.712, P <.001).

Similar questions were asked in reference to muscle strain: would reassurance be enough? Again, African Americans were less likely to see reassurance as sufficient (55% vs 81%, X²=14.003, P <.000). Seventy-two percent of those responding felt this would be sufficient care. However, 37% stated that they would like muscle relaxers or pain medication just to be sure. Patients who were in an HMO and African Americans were more likely to prefer the medication than those not in an HMO (55% HMO vs 36% non-HMO, X²=6.014, P<.014; 63% African Americans vs 37% whites, X²=9.59, P<.002).

Finally, the same questions were asked about diarrhea, which was most likely caused by a virus. Reassurance was again less apt to be perceived as sufficient by African Americans (50% vs 65%, X²= 3.442, P<.064), and they were more likely to want an antibiotic “just to be sure” (37% vs 23%, X²=3.759, P<.053). The patients not in an HMO were also more likely to want an antibiotic just to be sure (35% vs 21%, X²=4.669, P<.031).

Treatment Change Questions

Two questions were asked about the timing of requests for different treatments. The first had to do with the timing of a change in medicine (assuming the patient was not getting sicker) and the second with the number of medication changes that would be acceptable before requesting a referral to a specialist. Overall, 14% of the respondents would call for new medicine within 24 hours; 22% would call within 48 hours; 31% would call within 72 hours; and 33% would wait longer. In terms of requesting a specialist, 8% of the respondents would accept one medicine change before calling a specialist; 43% would accept 2 changes; 36% would accept 3 changes; and 13% would accept more changes. African Americans and FFS patients would wait for a shorter amount of time before calling for a medication change (F=14.66, P <017; F=6.34, P <013, respectively). No differences were observed by sex, race, or HMO membership status in number of medication changes before going to a specialist.

Discussion

We addressed patient preferences concerning 3 main topics: (1) influence of cost of medicine, (2) use of reassurance as a valid treatment, and (3) timing of treatment changes. Results indicate that patients want expensive medicine if there is a reason to believe it will work better. This finding is particularly true of HMO enrollees who have a set charge for medicine, suggesting that such plans and provisions may change patient preferences in a more costly manner or that patients self-select for membership. The acceptance of reassurance as a valid treatment by many patients belies the stereotype of patients always wanting prescriptions. However, variation in the wording of that question (the switch to whether they would want medication “just to be sure”) significantly reduced the percentage of those who simply accepted over-the-counter treatments and reassurance as adequate treatments. Substantial variation was observed by race, with African Americans consistently not as accepting of reassurance and more desirous of “just to be sure” medication. Although the small sample size precludes the drawing of conclusions, this may be suggestive of cultural differences in health-seeking behavior and also reduced exposure to managed care plans. One possibility is that African Americans may have a higher threshold for consulting a physician and may have tried symptomatic treatment for a longer period of time than whites. Because FFS patients were more likely to be African American, observed differences by race may be because of the covariation of race with insurer type. Future studies should explore these possible explanations.

We also found that patients are willing to accept multiple changes in medicines before seeking specialist care and to wait several days before calling physicians about medication ineffectiveness. African Americans and FFS members were less willing to wait. Perhaps there is a self-selection factor to enrollment in HMOs, or perhaps HMO members change their expectations after enrollment. In any case, these data show that understanding the role of patient preferences and expectations about the care they receive may have some potential for cost saving through the use of reassurance and corresponding patient satisfaction with care that does not always deliver a prescription.

Acknowledgements

The authors wish to thank Pam Stovall and Tracey Barton for their typing of the manuscript.