Antibiotic Prescribing for Children With Nasopharyngitis (Common Colds), Upper Respiratory Infections, and Bronchitis Who Have Health-Profes
http://www.100md.com
《小儿科》
School of Medicine, National Yang Ming University, Taipei, Taiwan
Institute of Health Care and Hospital Administration, National Yang Ming University, Taipei, Taiwan
Department of Social Medicine, National Yang Ming University, Taipei, Taiwan
Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
Bureau of National Health Insurance, Taipei, Taiwan
ABSTRACT
Objective. Antibiotic resistance might be reduced if patients could be better informed regarding the lack of benefits of antibiotics for children with viral infections and avoid antibiotic prescriptions in these circumstances. This study investigated whether children having health professionals as parents, a group whose parents are expected to have more medical knowledge and expertise, are less likely than other children to receive antibiotics for nasopharyngitis (common colds), upper respiratory tract infections (URIs), and acute bronchitis.
Methods. Retrospective analyses were conducted by using National Health Insurance data for children of physicians, nurses, pharmacists, and non–health personnel, who had visited hospital outpatient departments or physician clinics for common colds, URIs, and acute bronchitis in Taiwan in 2000. A total of 53733 episodes of care for common colds, URIs, and acute bronchitis in a nationally representative sample of children (aged 18 years) living in nonremote areas were analyzed.
Results. The study found that, after adjusting for characteristics of the children (demographic, socioeconomic, and health status) and the treating physicians (demographic, practice style, and setting), children with a physician (odds ratio [OR]: 0.50; 95% confidence interval [CI]: 0.36–0.68) or a pharmacist (OR: 0.69; 95% CI: 0.52–0.91) as a parent were significantly less likely than other children to receive antibiotic prescriptions. The likelihood of receiving an antibiotic for the children of nurses (OR: 0.91; 95% CI: 0.77–1.09) was similar to that for children in the comparison group.
Conclusions. This finding supports our hypothesis that better parental education does help to reduce the frequency of injudicious antibiotic prescribing. Medical knowledge alone, however, may not fully reduce the overuse of antibiotics. Physician-parents, the expected medically savvy parents, can serve as a benchmark for the improvement potentially achievable in Taiwan through a combination of educational, regulatory, communication, and policy efforts targeted at more appropriate antibiotic prescribing in ambulatory settings.
Key Words: antibiotic prescribing children health-professional parents Taiwan
Abbreviations: URI, upper respiratory infection NHI, National Health Insurance ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification SES, socioeconomic status OR, odds ratio CI, confidence interval
The rapid spread of antibiotic resistance in community pathogens has underscored the urgency for reducing unnecessary antibiotic use.1–3 The overprescription of antimicrobial drugs in the ambulatory care setting can lead to adverse effects, development of bacterial resistance, and financial costs to both the patient and society.4–8 Mounting evidence suggests that the benefit from antibiotics is minimal for diseases with a primarily viral etiology.9–16 American, Asian, and European studies indicate that between 17% and 90% of children consulting for the common cold, upper respiratory tract infections (URIs), and acute bronchitis received antibiotic prescriptions.17–21 Of all the factors leading to antibiotic prescribing, a parent’s demand for antibiotics has often been mentioned as key.18,22–28 A parent’s demand for antibiotics may stem from a misunderstanding of the cause of viral respiratory infections or from a lack of awareness of current medical opinion regarding appropriate treatment modalities. Therefore, many believe that changing the misconceptions of parents regarding the effectiveness of antibiotics for children with viral infections could lead to a reduction in inappropriate antibiotic prescribing and the costs associated with that practice.29,30
However, whether and to what extent educational efforts targeting patients can effectively and safely reduce unnecessary antibiotic use in ambulatory care remains unclear.31–33 Three groups of parents that we can presume are well educated in medical matters are physicians, pharmacists, and nurses. Health-professional parents are expected to have more knowledge about the lack of benefit with antibiotic use in viral infections and the concern about antibiotic resistance with injudicious use. Hence, if medical knowledge can decrease the demand for antibiotics, we hypothesized that children with a physician, pharmacist, or nurse as a parent would be less likely to receive antibiotics for common colds, URIs, bronchitis, and bronchiolitis when compared with other children. In addition, previous findings have indicated that physicians may treat their peers, and the relatives of their peers, differently than their other patients, because the treating physicians may be under stronger pressure from these informed patients.34–36 They may also face greater medical scrutiny from more-informed parents for treatments of questionable medical value.
Furthermore, the care-seeking behavior and utilization patterns of the families of health personnel may provide useful insights into defining the appropriate use of health care and the quality of that care, because they represent medically savvy consumers and are familiar with the health care system.37,38 However, only limited information regarding the use of medical services by health personnel and their families is available.37–45 Of this limited literature, a few articles have been published concerning the children of physicians.34,37,46 In 1996, Diekema and colleagues found that children with a physician for a parent received preferential treatment in hospital emergency departments34 and were less likely than other children to use the pediatric emergency department for nonurgent problems.46 Because a physician is believed to be the medically savvy parent with greater power and ability to communicate with treating physicians and better access to care, the frequency of antibiotics overuse among physicians’ children should be significantly lower than other children in ambulatory care settings. The objective of this study is to investigate whether and to what extent physicians’ prescribing practices differ for children with a physician, pharmacist, or nurse for a parent in comparison to other children after financial barriers to treatment have been removed under a national health insurance program.
METHODS
Data Sources
In Taiwan, the National Health Insurance (NHI) program provides mandatory universal health insurance, offering comprehensive medical care coverage to all civilian Taiwanese residents. The NHI sample files, constructed and managed by the National Health Research Institute, consist of comprehensive utilization and enrollment information for a random sample of 185889 NHI beneficiaries out of a population of 21400826 enrollees throughout Taiwan in 2000. A multistage stratified systematic sampling design according to the degree of urbanization, geographic location, and administrative boundaries was used by the National Health Research Institute to select a representative sample. Details of the design and sampling scheme have been reported elsewhere.47,48 The NHI sample files include the NHI enrollment files, the NHI claims data, the NHI medical-personnel registry, and the hospital/clinic registry.
NHI enrollment and payment of premiums is mainly through payroll deduction for people with well-defined monthly wages and through individual tax payments for farmers, fishermen, and people without a well-defined monthly wage. The enrollment file tracks each beneficiary’s insurance status (the insured or the dependent) and the policy holder’s insurable wages and occupation. By law, the dependents of the insured are generally the insured’s spouse, lineal ascendants (parents/grandparents) who are unemployed, or lineal blood descendants within a second degree of relationship (children/grandchildren) who are either <20 years of age and not employed or are >20 years of age but incapable of making a living, including those who are attending school without employment.
Relevant fields in the NHI claims data include diagnosis, date of medical service, drugs prescribed and filled, dispensing method and anonymous identifiers of the patient, the hospital/clinic, and the physician providing the service. The NHI claims data also have the birth date and gender of each patient recorded. The NHI medical-personnel registry maintains data on each medical professional in Taiwan and records the type of medical profession, birth date, gender, and specialty. The NHI hospital registry maintains data on each medical institution in Taiwan and provides information on certification level and geographic location. Unique but anonymous personnel and hospital identifiers were used to link the claims data to these provider registries.
Sample
There were 51185 children who were 18 years of age in this nationally representative sample. The study cohort was limited to 38469 nonaboriginal children with a physician, pharmacist, nurse, or non–medical personnel as a parent who lived in a nonremote area and who had received a primary diagnosis of nasopharyngitis (ie, the common cold; International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 460), URI not otherwise specified (ICD-9-CM codes 465), or acute or unspecified bronchitis or bronchiolitis (ICD-9-CM codes 466 and 490) in 2000. To make the children in the comparison group (children with non–medical personnel as parents) more comparable with children with a medical-personnel parent, we only included children who had regular wage earners as parents such as civil servants, government employees, private-sector employees, teachers, employers, and professionals.
Subsequently, according to previous literature,2,49 we excluded 4866 visits (2.7%) that involved a concomitant diagnosis with other common outpatient infections that might be appropriately treated with an antibiotic. We used the exclusion criteria that are listed in the Appendix. We then excluded a small number (68 visits [0.04%]) of children who were treated by their own physician-parent. Sensitivity analyses were conducted for different exclusion criteria, and the results remained robust.
Of the remaining 173728 visits made by 21383 children in the cohort, we defined and constructed an episode of care and "window period" by following the ambulatory episodes of care methodology.50 A window period of 30 days was assigned to each of the conditions as the maximum amount of time for which follow-up of the condition would be reasonable.50,51 The remaining data included 53733 episodes and 21383 children. For each patient, only the first visit within each episode was analyzed. Because each parent-child pair may have had >1 episode in a year, the results were also weighted by the number of episodes per each parent-child pair so that the physician prescribing behavior for each parent-child pair during the year was weighted equally.
Variables
Information on the prescribing of antibiotics was obtained from NHI claims data. The dependent variable in this study was presence or absence of at least 1 antibiotic prescription for each new case of the common cold, URI, and bronchitis. It was modeled as a binary variable (0 = no antibiotic prescription; 1 = antibiotic prescription). Topical antibiotics were excluded from this study.
The identification numbers of the primary insurance holders listed on the children’s enrollment files were matched against the NHI medical-personnel files to identify all children during the study period whose parents had the recorded occupation of "physician." Using the same files, we identified 2 other groups for which "nurse" and "pharmacist" were recorded as the occupation of the insured. We assumed that the insured person registered for each child dependent in the enrollment files was his/her parent.
Because the socioeconomic status (SES) and health status of physician-child pairs may differ from those in the nurse, pharmacist, and other nonmedical profession groups, we included SES and health status as control variables in the regression model. SES was inferred by linking the patient identifier and birth date to the NHI enrollment files and was defined by the parent’s insurable wages. As noted previously, NHI enrollment is mainly through payroll deduction for people with well-defined monthly wages. People with a well-defined monthly wage were classified into 3 categories: US $1280 and above, US $640 to US $1279, and less than US $640. The resource utilization band of the Johns Hopkins University Adjusted Clinical Groups Case-Mix System (version 6.0) was used to measure each patient’s comorbidity level.52 Three diagnosis fields in the NHI claims file covering all patient visits for 2000 were used to construct each individual’s adjusted-clinical-groups assignment, and then the resource-utilization-band algorithm assigned each individual to 1 of 4 morbidity levels: healthy, low morbidity, moderate morbidity, and high to very high morbidity.
Both simple and multiple logistic-regression models were conducted. First, a series of simple logistic models were fitted separately for each patient (age, gender, SES, comorbidity, or primary diagnosis of the visit) and physician (age, gender, specialty, certification level of practicing facility, dispensing method, or geographic area of practice) variable. Then, a multiple logistic model was fitted for all patient and physician variables. Seasonal variables were also included in the multiple logistic model as a possible confounder. The analyses were all adjusted for robust standard errors. A significance level of = .05 and the power of .80 were used. All analyses were conducted by using SAS 8.2 (SAS Institute, Cary, NC) and Stata 8.0 (Stata Corp, College Station, TX) statistical software packages. All individual patient and provider identifiers in the files used were encrypted. The confidentiality assurances were addressed by abiding the data regulations of the Bureau of National Health Insurance, and the data were used only for this research. This research was approved by the Johns Hopkins Bloomberg School of Public Health Committee on Human Research and the Bureau of National Health Insurance in Taiwan.
RESULTS
After adjusting for important children’s and treating physician’s confounders, physicians’ children (OR: 0.50; 95% CI: 0.36–0.68) were the least likely to receive antibiotics among all of the children, whether having nonphysician medical personnel or non–medical personnel as parents. The next least likely to receive an antibiotic prescription were children with pharmacist parents (OR: 0.69; 95% CI: 0.52–0.91). Unlike the other 2 medical-personnel groups, although children with nurse parents (OR: 0.91; 95% CI: 0.77–1.09) were less likely than children with non–medical personnel as parents to receive antibiotics, the difference did not reach statistical significance after taking all other influencing factors into consideration.
As for other characteristics, preschool children (OR: 1.23; 95% CI: 1.14–1.32) were more likely than other age groups to receive antibiotics after adjusting for parental occupation and other characteristics of the children and treating physicians. Children with parents whose monthly wages were less than US $640 had a significantly higher probability (OR: 1.03; 95% CI: 1.00–1.06) of receiving antibiotics than those with parents whose monthly wages were higher than US $1280. Children with low (OR: 0.93; 95% CI: 0.90–0.97) or moderate (OR: 0.91; 95% CI: 0.88–0.95) comorbidity levels had lower antibiotic prescribing. In terms of physician characteristics, older physicians and nonpediatricians were significantly more likely to prescribe antibiotics to children with common colds, URIs, or bronchitis during their first encounters after adjusting for the parents’, children’s, and other physicians’ characteristics. Physicians practicing in local hospitals and clinics were 3.19 times more likely to prescribe antibiotics to children than those practicing in medical centers and regional hospitals. Unlike their counterparts in Western countries, physicians in Taiwan are allowed to dispense drugs themselves or through their own on-site pharmacists. Prescriptions do not have to be dispensed by off-site pharmacists. Treating physicians who dispensed drugs or had their own on-site dispensing pharmacists were more than twice as likely to prescribe antibiotics as nondispensing doctors. The results suggest that a separation policy of prescribing and dispensing may serve as an additional modality to decrease inappropriate prescriptions. Antibiotic prescription variations resulting from geographical location of physician practices remained significant after adjusting for parental occupation and children’s and physicians’ characteristics.
DISCUSSION
These results suggest that among all children, those with physician-parents were least likely to receive potentially inappropriate antibiotic prescriptions, and their chance of being prescribed antibiotics with questionable value was significantly lower than children in general. This finding supports our hypothesis that children whose parents possess medical expertise are less likely to receive potentially inappropriate prescriptions and, in this regard, are more likely to receive better quality of care. However, changing the prescribing of antibiotics for all children to that of children having a physician as a parent will require considerable educational effort.
Apart from the children of physicians and pharmacists, although children of nurses who also possess extensive medical knowledge received antibiotics for nasopharyngitis, URIs, and bronchitis less frequently than the comparison children, the difference was not statistically significant. Hence, our findings suggest that this medical-knowledge factor alone would not assure us of a substantial reduction in the prescribing of antibiotics with questionable value. Similar to the findings of Diekema et al46 for emergency utilization, medical knowledge alone may not change the prescribing pattern of antibiotics.
There are 3 other possible explanations for the observed differences among the 3 categories of medical-personnel parents included in our study. First, the observed patterns may reflect different levels of perceived access to appropriate care if the child’s condition has been misdiagnosed or deteriorates.37,46 Because physician-parents often have special access to other physicians or are able to treat their own children, if necessary, they may show less concern or anxiety than pharmacists or nurses. Second, treating physicians may be more cautious in prescribing drugs to a child having a physician or pharmacist as a parent, because they are more likely to be under greater pressure and scrutiny from these 2 groups than from parents who are nurses.34–36 Hence, even if parental education does help to reduce the frequency of potentially inappropriate antibiotic prescribing, depending on this strategy alone will not be practical. Physician-parents, the informed consumers, may serve as a benchmark regarding the possible improvement that is feasible for antibiotic prescribing and the goal that may be reached through a combination of educational, regulatory, and policy efforts in Taiwan. Third, other than medical knowledge, because a physician-parent or spouse of a physician-parent might be more likely to have shared experiences and values, which would make communication easier, better communication between them and the treating physician could, in turn, lead to reduced antibiotic prescribing.37 Because physicians’ overestimation of parental expectation for antibiotics is cited often as a main factor of antibiotic prescription, better communication between treating physicians and health-personnel parents minimize this misconception.
Another observation is also worth mentioning. The frequency of antibiotic prescribing among the children of physicians (28%), the most well-informed medical care consumers in Taiwan, is still substantially higher than the average prescribing rates in the Netherlands (17%) and some other Western countries. The reasons behind this observation remain to be explored. Cultural factors may play an important role, and whether these cultural factors can be modified or changed remains uncertain. With this in mind, policy makers should be cautious when conducting international comparisons and establishing an absolute global gold standard of potentially inappropriate antibiotic prescribing. "One size fits all" may not be a realistic policy goal for global efforts in reducing the excessive use of antibiotics in ambulatory care.
Several study limitations should be noted. First, misclassification may be a source of bias. Because the NHI premium is linked to the insured’s income level, some children of physicians are likely to be insured through the other parent who may be in a lower income bracket and thus pay a lower premium; hence, these children may belong under another group. If this is the case, the true difference between physicians’ and nonphysicians’ children would be even larger than the one that we observed. Second, another misclassification bias may come from misclassification of antibiotic prescription with principal diagnosis. The principal diagnosis may not be the main complaint for that visit, and assuming that the antibiotics prescribed were related to the primary diagnosis for that particular encounter may be imperfect, because the claim forms contain up to 3 diagnoses. Therefore, sensitivity analyses were conducted on visits for the common cold, URI, or bronchitis for which there was only a single diagnosis. The results remained robust. Third, this study might suffer from certain inherent limitations from the use of claims data containing limited diagnostic and clinical information. Chart review may be more informative in this regard. Fourth, physicians may treat their own children themselves and not necessarily file a claim to the NHI program for reimbursement. The study looks only at those children whose parents used NHI medical services when their children had common colds, URIs, or bronchitis and may not reflect the differences in antibiotic prescribing by the entire population of physician-parents and others. Fifth, because of data limitation, this study used the occupation of the insured parents as a proxy of the parent’s medical knowledge, but the insured parent might not be the accompanying parent who brings the child to the doctor, especially for physician-parents. This would suggest that the alternative hypotheses above better explain the findings than the hypothesis of medical knowledge. Future research with more detailed data will allow for a purer test of the "knowledge hypothesis." Sixth, generalizability of the results may be a concern. Because the study was conducted in a national health insurance setting with nearly universal access, the results may not apply to countries with different health care systems.
Widespread antibiotic prescriptions for primarily viral conditions are likely to continue despite efforts to educate parents. In our study, children of physicians still received antibiotics frequently for these viral diseases, suggesting that parental education alone is not likely to eliminate all prescribing of antibiotics with questionable value in ambulatory care settings. Antibiotic prescribing observed for the children of physicians can represent the best outcome achievable through parental education. In addition, the differences observed among the children of physicians, pharmacists, and nurses and the comparison children suggest that increasing medical knowledge may still be insufficient to effect change; factors other than parental education, such as ready access to care on an urgent basis, better communication between treating physicians and parents, and the physician-patient relationships when the patient is a physician, may play a role in reducing the inappropriate prescribing of antibiotics. Future studies need to explore not only education or regulation interventions but also interventions to train both patients and health care providers in communication skills.53 A combination of strategies may help to substantially reduce the potentially inappropriate prescribing of antibiotics for commonly seen viral respiratory infections among the pediatric population in ambulatory care.
APPENDIX. Exclusion Criteria
ACKNOWLEDGMENTS
This study was supported by National Scientific Council grant 94-2320-B-003-004.
We sincerely thank David Bishai, MD, PhD, Christopher Forrest, MD, PhD, and Peter Pronovost, MD, PhD, for helpful comments on previous drafts.
FOOTNOTES
Accepted May 23, 2005.
No conflict of interest declared.
REFERENCES
Kunin CM. Resistance to antimicrobial drugs—a worldwide calamity. Ann Intern Med. 1993;118 :557 –561
Steinman MA, Gonzales R, Linder JA, Landefeld CS. Changing use of antibiotics in community-based outpatient practice, 1991–1999. Ann Intern Med. 2003;138 :525 –533
Nash DR, Harman J, Wald ER, Kelleher KJ. Antibiotic prescribing by primary care physicians for children with upper respiratory tract infections. Arch Pediatr Adolesc Med. 2002;156 :1114 –1119
Gonzales R, Malone DC, Maselli JH, Sande MA. Excessive antibiotic use for acute respiratory infections in the United States. Clin Infect Dis. 2001;33 :757 –762
Livermore DM. Bacterial resistance: origins, epidemiology, and impact. Clin Infect Dis. 2003;36(suppl 1) :S11 –S23
Mainous AG III, Hueston WJ. The cost of antibiotics in treating upper respiratory tract infections in a Medicaid population. Arch Fam Med. 1998;7 :45 –49
Fraser GL, Stogsdill P, Dickens JD Jr, Wennberg DE, Smith RP Jr, Prato BS. Antibiotic optimization. An evaluation of patient safety and economic outcomes. Arch Intern Med. 1997;157 :1689 –1694 [published correction appears in Arch Intern Med. 1997;157:2487]
Swartz MN. Use of antimicrobial agents and drug resistance. N Engl J Med. 1997;337 :491 –492
Gonzales R, Bartlett JG, Besser RE, et al. Principles of appropriate antibiotic use for treatment of uncomplicated acute bronchitis: background. Ann Emerg Med. 2001;37 :720 –727
Gonzales R, Bartlett JG, Besser RE, Hickner JM, Hoffman JR, Sande MA. Principles of appropriate antibiotic use for treatment of nonspecific upper respiratory tract infections in adults: background. Ann Emerg Med. 2001;37 :698 –702
Gonzales R, Bartlett JG, Besser RE, et al. Principles of appropriate antibiotic use for treatment of acute respiratory tract infections in adults: background, specific aims, and methods. Ann Emerg Med. 2001;37 :690 –697
Snow V, Mottur-Pilson C, Gonzales R, et al. Principles of appropriate antibiotic use for treatment of acute bronchitis in adults. Ann Intern Med. 2001;134 :518 –520
Snow V, Mottur-Pilson C, Gonzales R, et al. Principles of appropriate antibiotic use for treatment of nonspecific upper respiratory tract infections in adults. Ann Intern Med. 2001;134 :487 –489
Mainous AG 3rd, Hueston WJ. Antibiotics for the common cold. Lancet. 1996;348 :754
Arroll B, Kenealy T. Antibiotics for the common cold. Cochrane Database Syst Rev. 2002;(3) :CD000247
Arroll B, Kenealy T. Antibiotics for acute bronchitis. BMJ. 2001;322 :939 –940
Nyquist AC, Gonzales R, Steiner JF, Sande MA. Antibiotic prescribing for children with colds, upper respiratory tract infections, and bronchitis. JAMA. 1998;279 :875 –877
Gonzales R, Sande M. What will it take to stop physicians from prescribing antibiotics in acute bronchitis Lancet. 1995;345 :665 –666
McCaig LF, Besser RE, Hughes JM. Trends in antimicrobial prescribing rates for children and adolescents. JAMA. 2002;287 :3096 –3102
Cho HJ, Hong SJ, Park S. Knowledge and beliefs of primary care physicians, pharmacists, and parents on antibiotic use for the pediatric common cold. Soc Sci Med. 2004;58 :623 –629
de Melker RA, Kuyvenhoven MM. Management of upper respiratory tract infections in Dutch family practice. J Fam Pract. 1994;38 :353 –357
Butler CC, Rollnick S, Kinnersley P, Jones A, Stott N. Reducing antibiotics for respiratory tract symptoms in primary care: consolidating ‘why’ and considering ‘how. ’ Br J Gen Pract. 1998;48 :1865 –1870
Schwartz B, Mainous AG 3rd, Marcy SM. Why do physicians prescribe antibiotics for children with upper respiratory tract infections JAMA. 1998;279 :881 –882
Stivers T, Mangione-Smith R, Elliott MN, McDonald L, Heritage J. Why do physicians think parents expect antibiotics What parents report vs what physicians believe. J Fam Pract. 2003;52 :140 –148
Mangione-Smith R, McGlynn EA, Elliott MN, McDonald L, Franz CE, Kravitz RL. Parent expectations for antibiotics, physician-parent communication, and satisfaction. Arch Pediatr Adolesc Med. 2001;155 :800 –806
Mangione-Smith R, McGlynn EA, Elliott MN, Krogstad P, Brook RH. The relationship between perceived parental expectations and pediatrician antimicrobial prescribing behavior. Pediatrics. 1999;103 :711 –718
Butler CC, Rollnick S, Pill R, Maggs-Rapport F, Stott N. Understanding the culture of prescribing: qualitative study of general practitioners’ and patients’ perceptions of antibiotics for sore throats. BMJ. 1998;317 :637 –642
Kuzujanakis M, Kleinman K, Rifas-Shiman S, Finkelstein JA. Correlates of parental antibiotic knowledge, demand, and reported use. Ambul Pediatr. 2003;3 :203 –210
Raisch DW. A model of methods for influencing prescribing: part II. A review of educational methods, theories of human inference, and delineation of the model. DICP. 1990;24 :537 –542
Raisch DW. A model of methods for influencing prescribing: part I. A review of prescribing models, persuasion theories, and administrative and educational methods. DICP. 1990;24 :417 –421
Gonzales R, Steiner JF, Lum A, Barrett PH Jr. Decreasing antibiotic use in ambulatory practice: impact of a multidimensional intervention on the treatment of uncomplicated acute bronchitis in adults. JAMA. 1999;281 :1512 –1519
Ramsay C, Brown E, Hartman G, Davey P. Room for improvement: a systematic review of the quality of evaluations of interventions to improve hospital antibiotic prescribing. J Antimicrob Chemother. 2003;52 :764 –771
Mainous AG 3rd, Hueston WJ, Love MM, Evans ME, Finger R. An evaluation of statewide strategies to reduce antibiotic overuse. Fam Med. 2000;32 :22 –29
Diekema DS, Cummings P, Quan L. Physicians’ children are treated differently in the emergency department. Am J Emerg Med. 1996;14 :6 –9
Schneck SA. "Doctoring" doctors and their families. JAMA. 1998;280 :2039 –2042
Marzuk PM. When the patient is a physician. N Engl J Med. 1987;317 :1409 –1411
Wasserman RC, Hassuk BM, Young PC, Land ML. Health care of physicians’ children. Pediatrics. 1989;83 :319 –322
Wachtel TJ, Wilcox VL, Moulton AW, Tammaro D, Stein MD. Physicians’ utilization of health care. J Gen Intern Med. 1995;10 :261 –265
Bunker JP, Brown BW Jr. The physician-patient as an informed consumer of surgical services. N Engl J Med. 1974;290 :1051 –1055
Kahn KL, Goldberg RJ, DeCosimo D, Dalen JE. Health maintenance activities of physicians and nonphysicians. Arch Intern Med. 1988;148 :2433 –2436
Frank E, Brogan DJ, Mokdad AH, Simoes EJ, Kahn HS, Greenberg RS. Health-related behaviors of women physicians vs other women in the United States. Arch Intern Med. 1998;158 :342 –348
Frank E, Rothenberg R, Lewis C, Belodoff BF. Correlates of physicians’ prevention-related practices. Findings from the Women Physicians’ Health Study. Arch Fam Med. 2000;9 :359 –367
Rosvold EO, Bjertness E. Illness behaviour among Norwegian physicians. Scand J Public Health. 2002;30 :125 –132
Rosvold EO, Hjartaker A, Bjertness E, Lund E. Breast self-examination and cervical cancer testing among Norwegian female physicians. A nation-wide comparative study. Soc Sci Med. 2001;52 :249 –258
Petrikovsky B, Cohen M, Andreoli JW Jr. Gynecologic screening examinations: does the obstetrician-gynecologist’s spouse comply Am J Obstet Gynecol. 1989;160 :82 –85
Diekema DS, Del Beccaro MA, Cummings P, Quan L. Physician parents and utilization of a pediatric emergency department. Pediatr Emerg Care. 1996;12 :400 –403
Shih YT, Hung YT, Chang HY, et al. The design, contents, operation and the characteristics of the respondents of the 2001 National Health Interview Survey in Taiwan. Taiwan J Public Health. 2003;22 :465 –473
Chang HY, Chiou CJ, Lin MC, Lin SH, Tai TY. A population study of the self-care behaviors and their associated factors of diabetes in Taiwan: results from the 2001 National Health Interview Survey in Taiwan. Prev Med. 2005;40 :344 –348
Steinman MA, Landefeld CS, Gonzales R. Predictors of broad-spectrum antibiotic prescribing for acute respiratory tract infections in adult primary care. JAMA. 2003;289 :719 –725
Forrest CB, Starfield B. The effect of first-contact care with primary care clinicians on ambulatory health care expenditures. J Fam Pract. 1996;43 :40 –48
Hueston WJ, Mainous AG 3rd, Brauer N, Mercuri J. Evaluation and treatment of respiratory infections: does managed care make a difference J Fam Pract. 1997;44 :572 –577
Johns Hopkins University School of Hygiene and Public Health. ACG Software Documentation and Users Manual. Version 6.0. Baltimore, MD: Johns Hopkins University; 2000
Roter DL, Hall JA. Doctors Talking With Patients/Patients Talking With Doctors: Improving Communication Skills in Medical Visits. 1st ed. Westport, CT: Auburn House; 1993(Nicole Huang, PhD, Laura )
Institute of Health Care and Hospital Administration, National Yang Ming University, Taipei, Taiwan
Department of Social Medicine, National Yang Ming University, Taipei, Taiwan
Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland
Bureau of National Health Insurance, Taipei, Taiwan
ABSTRACT
Objective. Antibiotic resistance might be reduced if patients could be better informed regarding the lack of benefits of antibiotics for children with viral infections and avoid antibiotic prescriptions in these circumstances. This study investigated whether children having health professionals as parents, a group whose parents are expected to have more medical knowledge and expertise, are less likely than other children to receive antibiotics for nasopharyngitis (common colds), upper respiratory tract infections (URIs), and acute bronchitis.
Methods. Retrospective analyses were conducted by using National Health Insurance data for children of physicians, nurses, pharmacists, and non–health personnel, who had visited hospital outpatient departments or physician clinics for common colds, URIs, and acute bronchitis in Taiwan in 2000. A total of 53733 episodes of care for common colds, URIs, and acute bronchitis in a nationally representative sample of children (aged 18 years) living in nonremote areas were analyzed.
Results. The study found that, after adjusting for characteristics of the children (demographic, socioeconomic, and health status) and the treating physicians (demographic, practice style, and setting), children with a physician (odds ratio [OR]: 0.50; 95% confidence interval [CI]: 0.36–0.68) or a pharmacist (OR: 0.69; 95% CI: 0.52–0.91) as a parent were significantly less likely than other children to receive antibiotic prescriptions. The likelihood of receiving an antibiotic for the children of nurses (OR: 0.91; 95% CI: 0.77–1.09) was similar to that for children in the comparison group.
Conclusions. This finding supports our hypothesis that better parental education does help to reduce the frequency of injudicious antibiotic prescribing. Medical knowledge alone, however, may not fully reduce the overuse of antibiotics. Physician-parents, the expected medically savvy parents, can serve as a benchmark for the improvement potentially achievable in Taiwan through a combination of educational, regulatory, communication, and policy efforts targeted at more appropriate antibiotic prescribing in ambulatory settings.
Key Words: antibiotic prescribing children health-professional parents Taiwan
Abbreviations: URI, upper respiratory infection NHI, National Health Insurance ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification SES, socioeconomic status OR, odds ratio CI, confidence interval
The rapid spread of antibiotic resistance in community pathogens has underscored the urgency for reducing unnecessary antibiotic use.1–3 The overprescription of antimicrobial drugs in the ambulatory care setting can lead to adverse effects, development of bacterial resistance, and financial costs to both the patient and society.4–8 Mounting evidence suggests that the benefit from antibiotics is minimal for diseases with a primarily viral etiology.9–16 American, Asian, and European studies indicate that between 17% and 90% of children consulting for the common cold, upper respiratory tract infections (URIs), and acute bronchitis received antibiotic prescriptions.17–21 Of all the factors leading to antibiotic prescribing, a parent’s demand for antibiotics has often been mentioned as key.18,22–28 A parent’s demand for antibiotics may stem from a misunderstanding of the cause of viral respiratory infections or from a lack of awareness of current medical opinion regarding appropriate treatment modalities. Therefore, many believe that changing the misconceptions of parents regarding the effectiveness of antibiotics for children with viral infections could lead to a reduction in inappropriate antibiotic prescribing and the costs associated with that practice.29,30
However, whether and to what extent educational efforts targeting patients can effectively and safely reduce unnecessary antibiotic use in ambulatory care remains unclear.31–33 Three groups of parents that we can presume are well educated in medical matters are physicians, pharmacists, and nurses. Health-professional parents are expected to have more knowledge about the lack of benefit with antibiotic use in viral infections and the concern about antibiotic resistance with injudicious use. Hence, if medical knowledge can decrease the demand for antibiotics, we hypothesized that children with a physician, pharmacist, or nurse as a parent would be less likely to receive antibiotics for common colds, URIs, bronchitis, and bronchiolitis when compared with other children. In addition, previous findings have indicated that physicians may treat their peers, and the relatives of their peers, differently than their other patients, because the treating physicians may be under stronger pressure from these informed patients.34–36 They may also face greater medical scrutiny from more-informed parents for treatments of questionable medical value.
Furthermore, the care-seeking behavior and utilization patterns of the families of health personnel may provide useful insights into defining the appropriate use of health care and the quality of that care, because they represent medically savvy consumers and are familiar with the health care system.37,38 However, only limited information regarding the use of medical services by health personnel and their families is available.37–45 Of this limited literature, a few articles have been published concerning the children of physicians.34,37,46 In 1996, Diekema and colleagues found that children with a physician for a parent received preferential treatment in hospital emergency departments34 and were less likely than other children to use the pediatric emergency department for nonurgent problems.46 Because a physician is believed to be the medically savvy parent with greater power and ability to communicate with treating physicians and better access to care, the frequency of antibiotics overuse among physicians’ children should be significantly lower than other children in ambulatory care settings. The objective of this study is to investigate whether and to what extent physicians’ prescribing practices differ for children with a physician, pharmacist, or nurse for a parent in comparison to other children after financial barriers to treatment have been removed under a national health insurance program.
METHODS
Data Sources
In Taiwan, the National Health Insurance (NHI) program provides mandatory universal health insurance, offering comprehensive medical care coverage to all civilian Taiwanese residents. The NHI sample files, constructed and managed by the National Health Research Institute, consist of comprehensive utilization and enrollment information for a random sample of 185889 NHI beneficiaries out of a population of 21400826 enrollees throughout Taiwan in 2000. A multistage stratified systematic sampling design according to the degree of urbanization, geographic location, and administrative boundaries was used by the National Health Research Institute to select a representative sample. Details of the design and sampling scheme have been reported elsewhere.47,48 The NHI sample files include the NHI enrollment files, the NHI claims data, the NHI medical-personnel registry, and the hospital/clinic registry.
NHI enrollment and payment of premiums is mainly through payroll deduction for people with well-defined monthly wages and through individual tax payments for farmers, fishermen, and people without a well-defined monthly wage. The enrollment file tracks each beneficiary’s insurance status (the insured or the dependent) and the policy holder’s insurable wages and occupation. By law, the dependents of the insured are generally the insured’s spouse, lineal ascendants (parents/grandparents) who are unemployed, or lineal blood descendants within a second degree of relationship (children/grandchildren) who are either <20 years of age and not employed or are >20 years of age but incapable of making a living, including those who are attending school without employment.
Relevant fields in the NHI claims data include diagnosis, date of medical service, drugs prescribed and filled, dispensing method and anonymous identifiers of the patient, the hospital/clinic, and the physician providing the service. The NHI claims data also have the birth date and gender of each patient recorded. The NHI medical-personnel registry maintains data on each medical professional in Taiwan and records the type of medical profession, birth date, gender, and specialty. The NHI hospital registry maintains data on each medical institution in Taiwan and provides information on certification level and geographic location. Unique but anonymous personnel and hospital identifiers were used to link the claims data to these provider registries.
Sample
There were 51185 children who were 18 years of age in this nationally representative sample. The study cohort was limited to 38469 nonaboriginal children with a physician, pharmacist, nurse, or non–medical personnel as a parent who lived in a nonremote area and who had received a primary diagnosis of nasopharyngitis (ie, the common cold; International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 460), URI not otherwise specified (ICD-9-CM codes 465), or acute or unspecified bronchitis or bronchiolitis (ICD-9-CM codes 466 and 490) in 2000. To make the children in the comparison group (children with non–medical personnel as parents) more comparable with children with a medical-personnel parent, we only included children who had regular wage earners as parents such as civil servants, government employees, private-sector employees, teachers, employers, and professionals.
Subsequently, according to previous literature,2,49 we excluded 4866 visits (2.7%) that involved a concomitant diagnosis with other common outpatient infections that might be appropriately treated with an antibiotic. We used the exclusion criteria that are listed in the Appendix. We then excluded a small number (68 visits [0.04%]) of children who were treated by their own physician-parent. Sensitivity analyses were conducted for different exclusion criteria, and the results remained robust.
Of the remaining 173728 visits made by 21383 children in the cohort, we defined and constructed an episode of care and "window period" by following the ambulatory episodes of care methodology.50 A window period of 30 days was assigned to each of the conditions as the maximum amount of time for which follow-up of the condition would be reasonable.50,51 The remaining data included 53733 episodes and 21383 children. For each patient, only the first visit within each episode was analyzed. Because each parent-child pair may have had >1 episode in a year, the results were also weighted by the number of episodes per each parent-child pair so that the physician prescribing behavior for each parent-child pair during the year was weighted equally.
Variables
Information on the prescribing of antibiotics was obtained from NHI claims data. The dependent variable in this study was presence or absence of at least 1 antibiotic prescription for each new case of the common cold, URI, and bronchitis. It was modeled as a binary variable (0 = no antibiotic prescription; 1 = antibiotic prescription). Topical antibiotics were excluded from this study.
The identification numbers of the primary insurance holders listed on the children’s enrollment files were matched against the NHI medical-personnel files to identify all children during the study period whose parents had the recorded occupation of "physician." Using the same files, we identified 2 other groups for which "nurse" and "pharmacist" were recorded as the occupation of the insured. We assumed that the insured person registered for each child dependent in the enrollment files was his/her parent.
Because the socioeconomic status (SES) and health status of physician-child pairs may differ from those in the nurse, pharmacist, and other nonmedical profession groups, we included SES and health status as control variables in the regression model. SES was inferred by linking the patient identifier and birth date to the NHI enrollment files and was defined by the parent’s insurable wages. As noted previously, NHI enrollment is mainly through payroll deduction for people with well-defined monthly wages. People with a well-defined monthly wage were classified into 3 categories: US $1280 and above, US $640 to US $1279, and less than US $640. The resource utilization band of the Johns Hopkins University Adjusted Clinical Groups Case-Mix System (version 6.0) was used to measure each patient’s comorbidity level.52 Three diagnosis fields in the NHI claims file covering all patient visits for 2000 were used to construct each individual’s adjusted-clinical-groups assignment, and then the resource-utilization-band algorithm assigned each individual to 1 of 4 morbidity levels: healthy, low morbidity, moderate morbidity, and high to very high morbidity.
Both simple and multiple logistic-regression models were conducted. First, a series of simple logistic models were fitted separately for each patient (age, gender, SES, comorbidity, or primary diagnosis of the visit) and physician (age, gender, specialty, certification level of practicing facility, dispensing method, or geographic area of practice) variable. Then, a multiple logistic model was fitted for all patient and physician variables. Seasonal variables were also included in the multiple logistic model as a possible confounder. The analyses were all adjusted for robust standard errors. A significance level of = .05 and the power of .80 were used. All analyses were conducted by using SAS 8.2 (SAS Institute, Cary, NC) and Stata 8.0 (Stata Corp, College Station, TX) statistical software packages. All individual patient and provider identifiers in the files used were encrypted. The confidentiality assurances were addressed by abiding the data regulations of the Bureau of National Health Insurance, and the data were used only for this research. This research was approved by the Johns Hopkins Bloomberg School of Public Health Committee on Human Research and the Bureau of National Health Insurance in Taiwan.
RESULTS
After adjusting for important children’s and treating physician’s confounders, physicians’ children (OR: 0.50; 95% CI: 0.36–0.68) were the least likely to receive antibiotics among all of the children, whether having nonphysician medical personnel or non–medical personnel as parents. The next least likely to receive an antibiotic prescription were children with pharmacist parents (OR: 0.69; 95% CI: 0.52–0.91). Unlike the other 2 medical-personnel groups, although children with nurse parents (OR: 0.91; 95% CI: 0.77–1.09) were less likely than children with non–medical personnel as parents to receive antibiotics, the difference did not reach statistical significance after taking all other influencing factors into consideration.
As for other characteristics, preschool children (OR: 1.23; 95% CI: 1.14–1.32) were more likely than other age groups to receive antibiotics after adjusting for parental occupation and other characteristics of the children and treating physicians. Children with parents whose monthly wages were less than US $640 had a significantly higher probability (OR: 1.03; 95% CI: 1.00–1.06) of receiving antibiotics than those with parents whose monthly wages were higher than US $1280. Children with low (OR: 0.93; 95% CI: 0.90–0.97) or moderate (OR: 0.91; 95% CI: 0.88–0.95) comorbidity levels had lower antibiotic prescribing. In terms of physician characteristics, older physicians and nonpediatricians were significantly more likely to prescribe antibiotics to children with common colds, URIs, or bronchitis during their first encounters after adjusting for the parents’, children’s, and other physicians’ characteristics. Physicians practicing in local hospitals and clinics were 3.19 times more likely to prescribe antibiotics to children than those practicing in medical centers and regional hospitals. Unlike their counterparts in Western countries, physicians in Taiwan are allowed to dispense drugs themselves or through their own on-site pharmacists. Prescriptions do not have to be dispensed by off-site pharmacists. Treating physicians who dispensed drugs or had their own on-site dispensing pharmacists were more than twice as likely to prescribe antibiotics as nondispensing doctors. The results suggest that a separation policy of prescribing and dispensing may serve as an additional modality to decrease inappropriate prescriptions. Antibiotic prescription variations resulting from geographical location of physician practices remained significant after adjusting for parental occupation and children’s and physicians’ characteristics.
DISCUSSION
These results suggest that among all children, those with physician-parents were least likely to receive potentially inappropriate antibiotic prescriptions, and their chance of being prescribed antibiotics with questionable value was significantly lower than children in general. This finding supports our hypothesis that children whose parents possess medical expertise are less likely to receive potentially inappropriate prescriptions and, in this regard, are more likely to receive better quality of care. However, changing the prescribing of antibiotics for all children to that of children having a physician as a parent will require considerable educational effort.
Apart from the children of physicians and pharmacists, although children of nurses who also possess extensive medical knowledge received antibiotics for nasopharyngitis, URIs, and bronchitis less frequently than the comparison children, the difference was not statistically significant. Hence, our findings suggest that this medical-knowledge factor alone would not assure us of a substantial reduction in the prescribing of antibiotics with questionable value. Similar to the findings of Diekema et al46 for emergency utilization, medical knowledge alone may not change the prescribing pattern of antibiotics.
There are 3 other possible explanations for the observed differences among the 3 categories of medical-personnel parents included in our study. First, the observed patterns may reflect different levels of perceived access to appropriate care if the child’s condition has been misdiagnosed or deteriorates.37,46 Because physician-parents often have special access to other physicians or are able to treat their own children, if necessary, they may show less concern or anxiety than pharmacists or nurses. Second, treating physicians may be more cautious in prescribing drugs to a child having a physician or pharmacist as a parent, because they are more likely to be under greater pressure and scrutiny from these 2 groups than from parents who are nurses.34–36 Hence, even if parental education does help to reduce the frequency of potentially inappropriate antibiotic prescribing, depending on this strategy alone will not be practical. Physician-parents, the informed consumers, may serve as a benchmark regarding the possible improvement that is feasible for antibiotic prescribing and the goal that may be reached through a combination of educational, regulatory, and policy efforts in Taiwan. Third, other than medical knowledge, because a physician-parent or spouse of a physician-parent might be more likely to have shared experiences and values, which would make communication easier, better communication between them and the treating physician could, in turn, lead to reduced antibiotic prescribing.37 Because physicians’ overestimation of parental expectation for antibiotics is cited often as a main factor of antibiotic prescription, better communication between treating physicians and health-personnel parents minimize this misconception.
Another observation is also worth mentioning. The frequency of antibiotic prescribing among the children of physicians (28%), the most well-informed medical care consumers in Taiwan, is still substantially higher than the average prescribing rates in the Netherlands (17%) and some other Western countries. The reasons behind this observation remain to be explored. Cultural factors may play an important role, and whether these cultural factors can be modified or changed remains uncertain. With this in mind, policy makers should be cautious when conducting international comparisons and establishing an absolute global gold standard of potentially inappropriate antibiotic prescribing. "One size fits all" may not be a realistic policy goal for global efforts in reducing the excessive use of antibiotics in ambulatory care.
Several study limitations should be noted. First, misclassification may be a source of bias. Because the NHI premium is linked to the insured’s income level, some children of physicians are likely to be insured through the other parent who may be in a lower income bracket and thus pay a lower premium; hence, these children may belong under another group. If this is the case, the true difference between physicians’ and nonphysicians’ children would be even larger than the one that we observed. Second, another misclassification bias may come from misclassification of antibiotic prescription with principal diagnosis. The principal diagnosis may not be the main complaint for that visit, and assuming that the antibiotics prescribed were related to the primary diagnosis for that particular encounter may be imperfect, because the claim forms contain up to 3 diagnoses. Therefore, sensitivity analyses were conducted on visits for the common cold, URI, or bronchitis for which there was only a single diagnosis. The results remained robust. Third, this study might suffer from certain inherent limitations from the use of claims data containing limited diagnostic and clinical information. Chart review may be more informative in this regard. Fourth, physicians may treat their own children themselves and not necessarily file a claim to the NHI program for reimbursement. The study looks only at those children whose parents used NHI medical services when their children had common colds, URIs, or bronchitis and may not reflect the differences in antibiotic prescribing by the entire population of physician-parents and others. Fifth, because of data limitation, this study used the occupation of the insured parents as a proxy of the parent’s medical knowledge, but the insured parent might not be the accompanying parent who brings the child to the doctor, especially for physician-parents. This would suggest that the alternative hypotheses above better explain the findings than the hypothesis of medical knowledge. Future research with more detailed data will allow for a purer test of the "knowledge hypothesis." Sixth, generalizability of the results may be a concern. Because the study was conducted in a national health insurance setting with nearly universal access, the results may not apply to countries with different health care systems.
Widespread antibiotic prescriptions for primarily viral conditions are likely to continue despite efforts to educate parents. In our study, children of physicians still received antibiotics frequently for these viral diseases, suggesting that parental education alone is not likely to eliminate all prescribing of antibiotics with questionable value in ambulatory care settings. Antibiotic prescribing observed for the children of physicians can represent the best outcome achievable through parental education. In addition, the differences observed among the children of physicians, pharmacists, and nurses and the comparison children suggest that increasing medical knowledge may still be insufficient to effect change; factors other than parental education, such as ready access to care on an urgent basis, better communication between treating physicians and parents, and the physician-patient relationships when the patient is a physician, may play a role in reducing the inappropriate prescribing of antibiotics. Future studies need to explore not only education or regulation interventions but also interventions to train both patients and health care providers in communication skills.53 A combination of strategies may help to substantially reduce the potentially inappropriate prescribing of antibiotics for commonly seen viral respiratory infections among the pediatric population in ambulatory care.
APPENDIX. Exclusion Criteria
ACKNOWLEDGMENTS
This study was supported by National Scientific Council grant 94-2320-B-003-004.
We sincerely thank David Bishai, MD, PhD, Christopher Forrest, MD, PhD, and Peter Pronovost, MD, PhD, for helpful comments on previous drafts.
FOOTNOTES
Accepted May 23, 2005.
No conflict of interest declared.
REFERENCES
Kunin CM. Resistance to antimicrobial drugs—a worldwide calamity. Ann Intern Med. 1993;118 :557 –561
Steinman MA, Gonzales R, Linder JA, Landefeld CS. Changing use of antibiotics in community-based outpatient practice, 1991–1999. Ann Intern Med. 2003;138 :525 –533
Nash DR, Harman J, Wald ER, Kelleher KJ. Antibiotic prescribing by primary care physicians for children with upper respiratory tract infections. Arch Pediatr Adolesc Med. 2002;156 :1114 –1119
Gonzales R, Malone DC, Maselli JH, Sande MA. Excessive antibiotic use for acute respiratory infections in the United States. Clin Infect Dis. 2001;33 :757 –762
Livermore DM. Bacterial resistance: origins, epidemiology, and impact. Clin Infect Dis. 2003;36(suppl 1) :S11 –S23
Mainous AG III, Hueston WJ. The cost of antibiotics in treating upper respiratory tract infections in a Medicaid population. Arch Fam Med. 1998;7 :45 –49
Fraser GL, Stogsdill P, Dickens JD Jr, Wennberg DE, Smith RP Jr, Prato BS. Antibiotic optimization. An evaluation of patient safety and economic outcomes. Arch Intern Med. 1997;157 :1689 –1694 [published correction appears in Arch Intern Med. 1997;157:2487]
Swartz MN. Use of antimicrobial agents and drug resistance. N Engl J Med. 1997;337 :491 –492
Gonzales R, Bartlett JG, Besser RE, et al. Principles of appropriate antibiotic use for treatment of uncomplicated acute bronchitis: background. Ann Emerg Med. 2001;37 :720 –727
Gonzales R, Bartlett JG, Besser RE, Hickner JM, Hoffman JR, Sande MA. Principles of appropriate antibiotic use for treatment of nonspecific upper respiratory tract infections in adults: background. Ann Emerg Med. 2001;37 :698 –702
Gonzales R, Bartlett JG, Besser RE, et al. Principles of appropriate antibiotic use for treatment of acute respiratory tract infections in adults: background, specific aims, and methods. Ann Emerg Med. 2001;37 :690 –697
Snow V, Mottur-Pilson C, Gonzales R, et al. Principles of appropriate antibiotic use for treatment of acute bronchitis in adults. Ann Intern Med. 2001;134 :518 –520
Snow V, Mottur-Pilson C, Gonzales R, et al. Principles of appropriate antibiotic use for treatment of nonspecific upper respiratory tract infections in adults. Ann Intern Med. 2001;134 :487 –489
Mainous AG 3rd, Hueston WJ. Antibiotics for the common cold. Lancet. 1996;348 :754
Arroll B, Kenealy T. Antibiotics for the common cold. Cochrane Database Syst Rev. 2002;(3) :CD000247
Arroll B, Kenealy T. Antibiotics for acute bronchitis. BMJ. 2001;322 :939 –940
Nyquist AC, Gonzales R, Steiner JF, Sande MA. Antibiotic prescribing for children with colds, upper respiratory tract infections, and bronchitis. JAMA. 1998;279 :875 –877
Gonzales R, Sande M. What will it take to stop physicians from prescribing antibiotics in acute bronchitis Lancet. 1995;345 :665 –666
McCaig LF, Besser RE, Hughes JM. Trends in antimicrobial prescribing rates for children and adolescents. JAMA. 2002;287 :3096 –3102
Cho HJ, Hong SJ, Park S. Knowledge and beliefs of primary care physicians, pharmacists, and parents on antibiotic use for the pediatric common cold. Soc Sci Med. 2004;58 :623 –629
de Melker RA, Kuyvenhoven MM. Management of upper respiratory tract infections in Dutch family practice. J Fam Pract. 1994;38 :353 –357
Butler CC, Rollnick S, Kinnersley P, Jones A, Stott N. Reducing antibiotics for respiratory tract symptoms in primary care: consolidating ‘why’ and considering ‘how. ’ Br J Gen Pract. 1998;48 :1865 –1870
Schwartz B, Mainous AG 3rd, Marcy SM. Why do physicians prescribe antibiotics for children with upper respiratory tract infections JAMA. 1998;279 :881 –882
Stivers T, Mangione-Smith R, Elliott MN, McDonald L, Heritage J. Why do physicians think parents expect antibiotics What parents report vs what physicians believe. J Fam Pract. 2003;52 :140 –148
Mangione-Smith R, McGlynn EA, Elliott MN, McDonald L, Franz CE, Kravitz RL. Parent expectations for antibiotics, physician-parent communication, and satisfaction. Arch Pediatr Adolesc Med. 2001;155 :800 –806
Mangione-Smith R, McGlynn EA, Elliott MN, Krogstad P, Brook RH. The relationship between perceived parental expectations and pediatrician antimicrobial prescribing behavior. Pediatrics. 1999;103 :711 –718
Butler CC, Rollnick S, Pill R, Maggs-Rapport F, Stott N. Understanding the culture of prescribing: qualitative study of general practitioners’ and patients’ perceptions of antibiotics for sore throats. BMJ. 1998;317 :637 –642
Kuzujanakis M, Kleinman K, Rifas-Shiman S, Finkelstein JA. Correlates of parental antibiotic knowledge, demand, and reported use. Ambul Pediatr. 2003;3 :203 –210
Raisch DW. A model of methods for influencing prescribing: part II. A review of educational methods, theories of human inference, and delineation of the model. DICP. 1990;24 :537 –542
Raisch DW. A model of methods for influencing prescribing: part I. A review of prescribing models, persuasion theories, and administrative and educational methods. DICP. 1990;24 :417 –421
Gonzales R, Steiner JF, Lum A, Barrett PH Jr. Decreasing antibiotic use in ambulatory practice: impact of a multidimensional intervention on the treatment of uncomplicated acute bronchitis in adults. JAMA. 1999;281 :1512 –1519
Ramsay C, Brown E, Hartman G, Davey P. Room for improvement: a systematic review of the quality of evaluations of interventions to improve hospital antibiotic prescribing. J Antimicrob Chemother. 2003;52 :764 –771
Mainous AG 3rd, Hueston WJ, Love MM, Evans ME, Finger R. An evaluation of statewide strategies to reduce antibiotic overuse. Fam Med. 2000;32 :22 –29
Diekema DS, Cummings P, Quan L. Physicians’ children are treated differently in the emergency department. Am J Emerg Med. 1996;14 :6 –9
Schneck SA. "Doctoring" doctors and their families. JAMA. 1998;280 :2039 –2042
Marzuk PM. When the patient is a physician. N Engl J Med. 1987;317 :1409 –1411
Wasserman RC, Hassuk BM, Young PC, Land ML. Health care of physicians’ children. Pediatrics. 1989;83 :319 –322
Wachtel TJ, Wilcox VL, Moulton AW, Tammaro D, Stein MD. Physicians’ utilization of health care. J Gen Intern Med. 1995;10 :261 –265
Bunker JP, Brown BW Jr. The physician-patient as an informed consumer of surgical services. N Engl J Med. 1974;290 :1051 –1055
Kahn KL, Goldberg RJ, DeCosimo D, Dalen JE. Health maintenance activities of physicians and nonphysicians. Arch Intern Med. 1988;148 :2433 –2436
Frank E, Brogan DJ, Mokdad AH, Simoes EJ, Kahn HS, Greenberg RS. Health-related behaviors of women physicians vs other women in the United States. Arch Intern Med. 1998;158 :342 –348
Frank E, Rothenberg R, Lewis C, Belodoff BF. Correlates of physicians’ prevention-related practices. Findings from the Women Physicians’ Health Study. Arch Fam Med. 2000;9 :359 –367
Rosvold EO, Bjertness E. Illness behaviour among Norwegian physicians. Scand J Public Health. 2002;30 :125 –132
Rosvold EO, Hjartaker A, Bjertness E, Lund E. Breast self-examination and cervical cancer testing among Norwegian female physicians. A nation-wide comparative study. Soc Sci Med. 2001;52 :249 –258
Petrikovsky B, Cohen M, Andreoli JW Jr. Gynecologic screening examinations: does the obstetrician-gynecologist’s spouse comply Am J Obstet Gynecol. 1989;160 :82 –85
Diekema DS, Del Beccaro MA, Cummings P, Quan L. Physician parents and utilization of a pediatric emergency department. Pediatr Emerg Care. 1996;12 :400 –403
Shih YT, Hung YT, Chang HY, et al. The design, contents, operation and the characteristics of the respondents of the 2001 National Health Interview Survey in Taiwan. Taiwan J Public Health. 2003;22 :465 –473
Chang HY, Chiou CJ, Lin MC, Lin SH, Tai TY. A population study of the self-care behaviors and their associated factors of diabetes in Taiwan: results from the 2001 National Health Interview Survey in Taiwan. Prev Med. 2005;40 :344 –348
Steinman MA, Landefeld CS, Gonzales R. Predictors of broad-spectrum antibiotic prescribing for acute respiratory tract infections in adult primary care. JAMA. 2003;289 :719 –725
Forrest CB, Starfield B. The effect of first-contact care with primary care clinicians on ambulatory health care expenditures. J Fam Pract. 1996;43 :40 –48
Hueston WJ, Mainous AG 3rd, Brauer N, Mercuri J. Evaluation and treatment of respiratory infections: does managed care make a difference J Fam Pract. 1997;44 :572 –577
Johns Hopkins University School of Hygiene and Public Health. ACG Software Documentation and Users Manual. Version 6.0. Baltimore, MD: Johns Hopkins University; 2000
Roter DL, Hall JA. Doctors Talking With Patients/Patients Talking With Doctors: Improving Communication Skills in Medical Visits. 1st ed. Westport, CT: Auburn House; 1993(Nicole Huang, PhD, Laura )