 |
|
Does Drug Treatment of Patients With Acute Bronchitis Reduce Additional Care Seeking?
Evidence From the Practice Partner Research Network
William J. Hueston, MD;
Ruth Jenkins, MS;
Arch G. Mainous, III, PhD
Arch Fam Med. 2000;9:997-1001.
ABSTRACT
| |
Background Considerable discussion has focused on treatment methods for patients with acute bronchitis.
Objective To examine whether antibiotic or bronchodilator treatment is associated with differences in follow-up visit rates for patients with acute bronchitis.
Methods A retrospective medical chart review was conducted for patients with a new episode of acute bronchitis over a 3-year period in the Practice Partner Research Network (29 248 episodes in 24 753 patients). Primary outcomes of interest were another visit in the next 14 days (early follow-up) or 15 to 28 days after initial treatment (late follow-up).
Results Antibiotics were used more commonly in younger patients (<18 years), whereas older patients (>65 years) were more likely to receive no treatment. Younger patients treated with antibiotics were less likely to return for an early follow-up visit, but no differences were seen in adults and older patients. Late follow-up rates were not affected by the initial treatment strategy. When patients did return for a follow-up visit, no new medication was prescribed to most (66% of younger patients and 78% of older adults). However, compared with patients who did not receive an antibiotic at their first visit, patients initially treated with an antibiotic were about 50% more likely to receive a new antibiotic at their second visit.
Conclusions Initial prescribing of an antibiotic reduces early follow-up for acute bronchitis in younger patients but seems to have no effect in adults. However, reductions in future follow-up visits might be outweighed by increases in antibiotic consumption because patients who return for a follow-up visit seem to receive additional antibiotic prescriptions.
INTRODUCTION
IN THE PAST several years, many studies have reported the benefits or lack of benefits of treatments for acute bronchitis, including several individual studies of antibiotic therapy for symptomatic improvement of acute bronchitis1 and summaries and meta-analyses combining the results of these studies.2-3 In addition, 3 studies4-6 have examined the effect of bronchodilators on the pulmonary functions and symptoms of patients with acute bronchitis. Despite evidence7 suggesting that bronchodilator treatment reduces cough by approximately 50% in patients treated for a week and that use of antibiotics has only marginal effectiveness, clinicians seem to favor antibiotics for treating patients with acute bronchitis and rarely use bronchodilators.
The search for effective treatments continues despite the knowledge that acute bronchitis is a self-limited condition and patients will improve without treatment, suggesting that treatments for acute bronchitis might be prescribed for reasons other than curing the acute illness. Treatment goals might include reducing symptom severity; improving patient satisfaction, function, and quality of life; or allaying patient anxiety about their condition. One way of assessing whether these goals have been met is to examine how often patients return for continued care of the same bronchitis episode.
Although previous studies have focused on a variety of clinical outcomes, few have examined whether treatment alters subsequent health care utilization during the episode of acute bronchitis. Although self-limited, acute bronchitis can be a fairly long-term problem, with up to one quarter of patients continuing to cough for a month or longer.8 In most clinical trials of medications, patients are required to follow-up with their physician for clinical assessments. This mandatory second visit makes it difficult to determine whether patients would have sought additional care for their condition after their initial visit. In a decision analysis of acute bronchitis,9 it was found that antibiotic treatment might be a cost-effective strategy if subsequent utilization is reduced in patients who receive antibiotics at the initial visit. However, treatment would have to reduce subsequent care seeking substantially to make this strategy cost beneficial.
The purpose of this study was to examine how often patients diagnosed as having acute bronchitis make subsequent physician visits during the 28 days after their initial diagnosis. In addition, we sought to determine whether initial treatment with an antibiotic vs a bronchodilator altered the rate of subsequent care seeking. Finally, we assessed what kinds of treatment were rendered for patients during follow-up visits to gauge how subsequent care seeking for acute bronchitis affected health care costs and drug use.
PATIENTS AND METHODS
This study was performed using data from the Practice Partner Research Network, a national network of practices that share a common electronic medical record system (Practice Partner; Physicians MicroSystems, Inc, Seattle, Wash) and combine data regarding diagnoses, preventive services, selected laboratory and radiology services, and prescriptions for research purposes. The Practice Partner Research Network includes 53 primary care practices and 4 specialty practices in 24 states in the United States. A self-running program collects data from participating practices each month that are then submitted to a central site for inclusion in a longitudinal database. The database includes more than 380 000 patients with data on 2.3 million outpatient contacts and 3.9 million diagnoses. For this study, the database containing information about patient visits between October 1, 1995, and September 30, 1998, was included for analysis. To ensure that patients in the study had not had a previous episode of bronchitis in the preceding 60 days and to allow for collection of follow-up data, patients were included if they presented with acute bronchitis between December 1, 1995, and September 1, 1998.
To identify patients with acute bronchitis we selected those with International Classification of Diseases, Ninth Revision, Clinical Modification10 code 466.0 (acute bronchitis). To ensure that this was a new episode of illness, we excluded any patient who had a diagnosis of acute bronchitis in the previous 60 days. We also excluded patients with other chronic respiratory conditions, including asthma, chronic bronchitis, emphysema, or chronic obstructive pulmonary disease, and patients with a history of congestive heart failure to ensure that the current symptoms represented a new acute illness rather than an exacerbation of a chronic condition. We also excluded all patients treated with both an antibiotic and a bronchodilator because it is unlikely that the use of these 2 agents are independent decisions but rather constitute a separate decision possibly for patients with more severe illness. Second, this strategy was used more than twice as often in younger patients (8.6% of visits) compared with older adults (3.7% of visits), which suggests that other factors affected the decision. These visits constituted only 6.7% (n = 1966) of the total episodes. After excluding these episodes, 27 282 remaining episodes of acute bronchitis were identified during the 21-month study.
Patient medical records were then examined to determine what type of treatment a patient received (antibiotic, bronchodilator, or neither) and whether the patient made a follow-up visit. We decided to divide return visits into early follow-up ( 14 days after treatment) and late follow-up (15-28 days after treatment). The rationale for this categorization is that patients are likely to be actively treated in the first couple of weeks after seeing the physician and a return visit at this point might signal a lack of faith that the treatment is working. Later follow-up is more likely to indicate that symptoms continue to occur after treatment has concluded. Any treatment rendered at follow-up also was recorded.
For analysis, patients were stratified into 3 age groups: younger than 18, 18 to 65, and older than 65 years. Follow-up rates were compared using  2 statistics for each stratum. Finally, to adjust for effects of sex and individual practice styles, a logistic regression analysis was performed using follow-up as the dependent variable and including practice, age, sex, treatment with an antibiotic, and treatment with a bronchodilator.
RESULTS
Distribution of patients by age and treatment is shown in Table 1. Younger patients were more likely to be treated initially with antibiotics compared with the other 2 groups (P<.001). Whereas bronchodilator use was relatively constant throughout the age strata, we found a positive relation between the use of no medication and age.
|
|
|
|
Table 1. Patient Age Distribution and Initial Treatment
|
|
|
We found that 13% of patients had early follow-up visits and 7% had late follow-up visits. Most patients returning for early follow-up visits (87%) or late follow-up visits (93%) made only 1 return visit after their diagnosis. Follow-up visits were related to patient age but not sex. The highest follow-up rates were found in patients older than 65 years: 18% of older patients had early follow-up visits and 11% had late follow-up visits. These rates in patients older than 65 years compare with early follow-up rates of 10% in patients younger than 18 years and 12% in 18- to 65-year-olds (P<.001 compared with >65-year-olds) and late follow-up rates of 7% in those younger than 18 years and 6% in 18- to 65-year-olds (P<.001).
We found little variation in the percentage of patients who returned for visits within 2 weeks and between 2 and 4 weeks based on initial treatment (Table 2). Younger patients who received antibiotics at their first visit were slightly less likely to return for a second visit in the next 2 weeks, but this was not observed in either of the 2 older patient strata. For all 3 age groups, there were no significant differences in late follow-up visit rates.
|
|
|
|
Table 2. Effects of Treatment on Follow-up in Patients With Acute Bronchitis
|
|
|
Logistic regression analyses found that after adjustment for age, sex, and practice, the odds of a return visit within 2 weeks decreased by 14% if treated with an antibiotic (odds ratio, 0.86; 95% confidence interval, 0.80-0.93). Antibiotic treatment was not associated with a difference in return visits between 15 and 28 days. Treatment with a bronchodilator was not a predictor for return visits in either time interval.
Finally, we examined what kind of treatment was rendered when patients returned for a follow-up visit. For all age groups, most patients received no new medication at the return visit. The percentage of patients receiving no new treatment ranged from 66% of patients younger than 18 years to 78% of patients older than 65 years. Nearly all other patients received an antibiotic. Patients in all 3 age groups who received an antibiotic at their initial visit were more likely to receive a second antibiotic at their return visit (Table 3).
|
|
|
|
Table 3. Treatment Rendered at Second Visit for Acute Bronchitis Based on Initial Treatment
|
|
|
COMMENT
The results of our study indicate that initial selection of an antibiotic for the treatment of acute bronchitis results in a small, but statistically significant, decrease in follow-up visits within the first 2 weeks of diagnosis. The reduction in visits is most pronounced in younger patients. However, use of an antibiotic or a bronchodilator seemed to have no effect on follow-up rates 15 to 28 days after diagnosis.
The potential beneficial effects of antibiotic treatment on reducing short-term utilization of additional services must be weighed against the cost and potential complications of prescribing more antibiotics. Simply on a case basis, it would be attractive to use antibiotics to reduce the costs associated with additional visits. However, this might be shortsighted because many more antibiotic prescriptions will be required to attain this reduction in visits. For example, in younger patients, who had the largest difference in follow-up rates for patients treated with antibiotics (9%) compared with no therapy (12%), the absolute risk reduction is 3%. This means that 33 (100/3) prescriptions for antibiotics would have to be written to prevent 1 return visit. Furthermore, 33% of those treated with an antibiotic who returned for second visits would receive a new antibiotic. So, for every 100 patients treated with an antibiotic initially, 103 total antibiotic prescriptions (100 initial plus 3 patients who make second visits) would be written for patients who received antibiotics at the first visit. This compares with a total of only about 3 antibiotic prescriptions for patients who received no treatment at their initial visit (none at the first visit plus 20% of the 13% of patients who made a second visit). So although younger patients who receive antibiotics might make fewer physician visits than those who do not initially receive an antibiotic, the cost of these additional visits must be compared with the cost of extra antibiotic prescriptions.
One perplexing question that continues to recur is "Why do clinicians continue to prescribe antibiotics for conditions for which there is little documented evidence that these drugs work?"9, 11 This study suggests that clinicians might use antibiotics because it keeps patients from coming back for return visits. The simple act of providing an antibiotic with a predefined duration of therapy might provide a window of opportunity for the condition to resolve without the clinician having to confront patients whose symptoms are lingering. For bronchodilators, the standard duration of therapy based on previous trials is 1 week. After that time, patients with unresolved symptoms may return. If antibiotics are prescribed for a longer duration, such as 10 days or 2 weeks, this would provide additional opportunity for the self-limited condition to resolve. This could explain why more patients given bronchodilators (which have some evidence of benefit) are more likely to have a return visit within 2 weeks than those who are given antibiotics.12
But is use of an antibiotic justified or can it be viewed as a harmless placebo? In addition to the cost incurred with antibiotic prescribing, the use of antibiotics for viral infections may increase the rate of antibiotic resistance. Data from Finland13 suggest that reductions in the use of certain drug types are associated with declines in resistance to the same class of drugs. Other data14 show that as antibiotic use increases, the rate of penicillin-resistant pneumococcal carriage in nasopharyngeal swabs tends to increase. These studies suggest that antibiotic prescribing has ramifications beyond the effects in a single patient and could influence community or regional antibiotic resistance patterns. Consequently, it may be harmful to view antibiotics as a harmless placebo even when small benefits in subsequent utilization are associated with use of these agents.
Although antibiotic therapy had marginal effects on utilization rates, we also noted that prescribing a bronchodilator did not seem to reduce future use of health care services. Because bronchodilators have been shown to reduce cough within 1 week of use,4-5 we hypothesized that fewer patients would require ongoing care. Previous studies have shown that about half of all patients are cough free after a week of therapy, whereas the other half are still coughing. The follow-up rates observed in this study are much lower than 50%, which is consistent with previous results. However, it is disappointing that patients treated with bronchodilators seem to return for subsequent care at the same rate as patients who do not receive drug treatment. This finding casts doubt on whether routine bronchodilator use is a cost-effective strategy for treating these patients. More study would be useful in determining whether routine bronchodilator use is a cost-beneficial approach.
These results must be interpreted in view of the limitations of our study design. First, we could not control for the physician's perception of illness severity. Although there are no data to indicate that one treatment strategy is better than another for more or less severe illness, physicians might choose specific treatment regimens depending on their perception of the severity of a patient's illness. For example, physicians might have prescribed bronchodilators for patients with significant wheezing or more severe illness, which would increase the expected follow-up rate in this group compared with others. Data from other studies,15 however, suggest that physician prescribing is not based on their perception of illness severity as much as on individual prescriber behaviors, with some physicians falling into "high-prescriber" and others into "low-prescriber" categories.
Second, follow-up visits might not always represent treatment failure or patient anxiety over continuation of symptoms. Physicians often schedule "routine" follow-up visits when concerned about a patient's well-being. This could explain the higher follow-up rates we observed in the older patient group. However, because we do not have access to all the medical notes from follow-up visits and only record the diagnosis made at the visit, we cannot ascertain the actual changes in symptoms or the true reason for a patient's visit.
Also, the Practice Partner Research Network captures data for visits made to member practices only. The ability of patients to seek care at emergency facilities or other physicians' offices might result in an underestimate of the true follow-up rate. However, because of the large number of individuals included in this data set, to substantially change the relative distributions of follow-up based on different treatment strategies, a large number of patients would have had to seek care elsewhere.
A final point that must be raised is that acute bronchitis is a diverse problem that might be defined differently by different clinicians.16-17 The heterogeneity in the diagnosis of acute bronchitis has led some researchers18 to suggest that this clinical syndrome is no more than a chest cold and should not be given a separate diagnosis. Because clinicians differ in their diagnosis of acute bronchitis, treatments that work well in patients that some clinicians diagnose as having acute bronchitis might not work as well in other patients with different clinical symptoms diagnosed as having acute bronchitis by another physician. This heterogeneity can obscure important differences in the management of this condition, but it also reflects the reality of primary care practice. Research in conditions that are not pure and controlled should be based on the practice beliefs and decisions made by active clinicians. Exclusion of patients who do not meet rigid criteria might make for purer science but could produce results that might not be able to be applied to most primary care patients.
In summary, we found that early and late follow-up visits for acute bronchitis occur in similar percentages of patients regardless of initial treatment. Early and late follow-up visits are age related, with patients older than 65 years making early and late visits at significantly higher rates. Although initial treatment had little effect on the rate of subsequent care seeking, initial treatment with antibiotics was associated with higher antibiotic use rates at subsequent visits. Further exploration of why patients return for care and whether certain subsets of patients are at higher risk for future visits might be useful to target therapy more appropriately. In addition, further research should continue on additional interventions that might ameliorate the symptoms of acute bronchitis.
AUTHOR INFORMATION
Accepted for publication August 20, 2000.
This work was partially supported by IMS, America, Philadelphia, Pa, through funds to support the Practice Partner Research Network.
Reprints: William J. Hueston, MD, Department of Family Medicine, Medical University of South Carolina, 295 Calhoun St, PO Box 250192, Charleston, SC 29425 (e-mail: huestowj{at}musc.edu).
From the Department of Family Medicine and the Center for Health Care Research, Medical University of South Carolina, Charleston.
REFERENCES
| |
1. Orr PH, Scherer K, Macdonald A, Moffatt MEK. Randomized placebo-controlled trials of antibiotics for acute bronchitis: a critical review of the literature. J Fam Pract. 1993;36:507-512.
ISI
| PUBMED
2. Smucny JJ, Becker LA, Glazier RH, McIsaac W. Are antibiotics effective treatment for acute bronchitis? a meta-analysis. J Fam Pract. 1998;47:453-460.
ISI
| PUBMED
3. Fahey T, Stocks N, Thomas T. Quantitative systematic review of randomised controlled trials comparing antibiotic with placebo for acute cough in adults. BMJ. 1998;316:906-910.
FREE FULL TEXT
4. Hueston WJ. Albuterol delivered by metered-dose inhaler to treat acute bronchitis. J Fam Pract. 1994;39:437-440.
ISI
| PUBMED
5. Hueston WJ. A comparison of albuterol and erythromycin for the treatment of acute bronchitis. J Fam Pract. 1991;33:476-480.
ISI
| PUBMED
6. Melbye H, Aasebo U, Straume B. Symptomatic effect of inhaled fenoterol in acute bronchitis: a placebo-controlled double-blind study. Fam Pract. 1991;8:216-222.
FREE FULL TEXT
7. Mainous III AG, Zoorob RJ, Hueston WJ. Current management of acute bronchitis in ambulatory care: the use of antibiotics and bronchodilators. Arch Fam Med. 1996;5:79-83.
FREE FULL TEXT
8. Williamson HA Jr. A randomized, controlled trial of doxycycline in the treatment of acute bronchitis. J Fam Pract. 1984;19:481-486.
ISI
| PUBMED
9. Hueston WJ. Antibiotics in acute bronchitis: neither cost effective nor "cough" effective. J Fam Pract. 1997;44:261-265.
ISI
| PUBMED
10. International Classification of Diseases, Ninth Revision, Clinical Modification. Washington, DC: Public Health Service, US Dept of Health and Human Services; 1988.
11. Mainous AG, Hueston WJ, Clark JR. Do some folks think there is a cure for the common cold? evidence of widespread use of antibiotics in ambulatory care. J Fam Pract. 1996;42:357-361.
ISI
| PUBMED
12. Hueston WJ, Mainous AG, Brauer N, Mercuri J. Evaluation and treatment of respiratory infections: does managed care make a difference? J Fam Pract. 1997;44:572-577.
ISI
| PUBMED
13. Seppala H, Klaukka T, Vuopio-Varkila J, et al. The effect of changes in the consumption of macrolide antibiotics on erythromycin resistance in group A streptococci in Finland. N Engl J Med. 1997;337:441-446.
FREE FULL TEXT
14. Arason VA, Kristinsson KG, Sigurdsson JA, Stefansdottir G, Molstad S, Gudmundsson S. Do antimicrobials increase the carriage rate of penicillin-resistant pneumococci in children? cross-sectional prevalence study. BMJ. 1996;313:387-391.
FREE FULL TEXT
15. Mainous III AG, Hueston WJ, Love MD. Antibiotics for colds in children: who are the high prescribers? Arch Pediatr Adolesc Med. 1998;152:349-352.
FREE FULL TEXT
16. Oeffinger KC, Snell LM, Foster BM, Panico KG, Archer RK. Diagnosis of acute bronchitis in adults: a national survey of family physicians. J Fam Pract. 1997;45:402-409.
ISI
| PUBMED
17. Dunlay J, Reinhardt R. Clinical features and treatment of acute bronchitis. J Fam Pract. 1984;18:719-722.
ISI
| PUBMED
18. Hueston WJ, Mainous III AG, Dacus E, Hopper J. Does acute bronchitis really exist? J Fam Pract. 2000;49:401-406.
ISI
| PUBMED
|
