- Email: [email protected]
Prospective Tracking of Pediatric Urology Consults: Knowing is Half the Battle
Prospective Tracking of Pediatric Urology Consults: Knowing is Half the Battle Emilie K. Johnson,* Christopher P. Filson, Gary J. Faerber, John M. Park, David A. Bloom and Julian Wan From the Department of Urology, Children’s Hospital Boston, Boston, Massachusetts (EKJ), and Department of Urology, University of Michigan, Ann Arbor, Michigan
Abbreviations and Acronyms CDB ⫽ consult database ED ⫽ emergency department HO ⫽ house officer NGB ⫽ neurogenic bladder Submitted for publication September 16, 2011. * Correspondence: Department of Urology, Children’s Hospital Boston, 300 Longwood Ave., HU 3rd Floor, Boston, Massachusetts 02115 (telephone: 734-678-1594; FAX: 617-730-0474; e-mail: [email protected]).
Purpose: The planned clinical activity of pediatric urologists has been well described. However, little is known about nonscheduled work (eg consultation requests). We describe the unplanned clinical activity of pediatric urologists at a high volume academic medical center. Materials and Methods: Demographic data regarding inpatient, operating room and emergency department pediatric urology consults were prospectively entered into an internal database. Consults from July 2008 through June 2010 underwent retrospective chart abstraction to identify reasons for consultation. Bivariate and multivariate statistics were used to evaluate 1) temporal trends in unplanned clinical activity, and 2) patient and service specific factors associated with whether a consult was billable (ie seen by attending physician within 24 hours). Results: During the study period 665 pediatric consults were obtained. Mean ⫾ SD patient age was 8.4 ⫾ 7.7 years. Nearly all consults were seen at the emergency department (51%) or the inpatient wards (47%). The most common primary diagnoses were infection, obstruction/hydronephrosis and neurogenic bladder. The number of consults per month decreased during the course of the academic year (r2 ⫽ 0.1422). Nearly three fourths of consults were eligible for billing. The factors associated with consult eligibility for billing included specific attending physician (p ⫽ 0.03), location (p ⬍0.0001) and house officer experience (p ⫽ 0.007). Conclusions: At our academic pediatric hospital we averaged nearly 1 unplanned pediatric urology consult per day. Several service specific factors (unrelated to patient diagnosis or acuity) were associated with whether the consult had the potential to generate revenue. Unplanned clinical activity is an important factor to consider when planning departmental funding, staffing and training. Key Words: academic medical centers; pediatrics; referral and consultation; urology department, hospital
1844
www.jurology.com
PEDIATRIC urology joined the list of 147 medical specialties and subspecialties certified by the American Board of Medical Specialties in 2008. The scope of practice for each of these fields is constantly changing in response to technological, socioeconomical and regulatory forces. To finance and organize
health care delivery and education appropriately, it is essential to understand the full scope of practice and workforce expectations. Unplanned clinical consultation requests are an important component of the scope of any physician or department practice. Much of the literature
0022-5347/12/1875-1844/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 187, 1844-1849, May 2012 Printed in U.S.A. DOI:10.1016/j.juro.2011.12.097
AND
RESEARCH, INC.
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
regarding unplanned inpatient consultation comes from specialties such as dermatology and palliative care.1– 4 However, these specialties are consultation driven, with clinical work flow and revenue streams that differ from urological practice. The urological literature regarding unplanned clinical activity is limited but includes an investigation of consultations for iatrogenic catheter injuries.5 That study suggests that understanding the events leading to urgent urological consultation could improve patient outcomes through procedural education. In pediatric urology clinical activity has been catalogued by individual case logs, by databases of third party payers and by the American Board of Urology.6 These sources capture the activity of pediatric urologists at the administrative level but lack the granularity necessary to understand the details of hospital work flow. Therefore, little is known about the nonscheduled work done by pediatric urologists. Understanding unplanned clinical activity has the potential to inform educational initiatives (which ultimately improve patient safety), reveal patterns of documentation for appropriate billing and assist with streamlining work flow for appropriate staffing.7 Despite these potential benefits, consultations to pediatric urologists from EDs, inpatient settings and operating rooms have not previously been well described. Thus, we attempted to understand better the unplanned clinical activities in our academic pediatric urology practice.
METHODS Consult Work Flow At our institution requests for urology consultation are made by contacting the resident on call, who then evaluates the case and staffs the case with a faculty member. There is no specific pediatric urology consult service. Although on call for similar amounts of time in the evenings, HO III (junior) residents tend to be responsible for fielding consult calls during the day, which often results in the junior residents evaluating more consults than senior residents. The consult is staffed with the primary urologist of the patient (if applicable and available) or with the pediatric urologist on call. If necessary, consults are staffed with faculty members who are immediately available, even if they are not technically on call. This cooperative approach can result in different faculty members being responsible for unequal numbers of consults. Once the resident responsible for the encounter completes the documentation, details regarding the case are entered into a CDB.
Data Source An institutional CDB was created in 2006 as a means of tracking consults for quality control, education and billing. The CDB uses a password protected commercial database program (Microsoft® Access) and is housed on a secure medical center network server. Each record notes
1845
the date of service, medical identification number, patient gender, urology resident physician, urology attending physician, location (ED, operating room, inpatient, other) and type of interaction. Type of interaction is based on timing of staffing and defined as either “Seen same day” (ie same calendar day), “Seen in 24 hours” (ie not the same calendar day but within 24 hours) or “Telephone only.” The “Seen in 24 hours” category captures patients who were evaluated by the resident in the ED, then admitted and seen by the attending on the wards the next day. Cases that are managed only over the phone are listed as “Telephone only.” These options were chosen because they determined what type of billing could be made by most payers, as only consults seen within 24 hours are considered billable. Data entry was standardized by using dropdown menus for each field and entered prospectively. After an initial year to test the basic concept, we began to track fully all consults beginning in July 2007. In 2009 we added 3 fields to categorize the primary, secondary and tertiary reasons for consultation, which are determined by the physician completing the consult entry. Monthly reports of consult statistics are circulated to all resident and staff physicians. Data entry began in 2006 but the analysis for this project did not begin until 2008, when the CDB was a consistent part of work flow. Thus, we estimate the number of missing consults to be minimal.
Data Collection and Inclusion Criteria The CDB was queried for all pediatric consults seen between July 2008 and June 2010. We defined a pediatric consult as a clinical consultation for any patient younger than 18 years or a patient 18 or older still under the care of a pediatric urologist for a pediatric urological diagnosis. All consults that met this definition underwent retrospective chart review to confirm the primary, secondary and tertiary reasons for consultation. For patients seen before July 2009, when the diagnosis fields were added to the database, the reasons for consultation were obtained through chart review only.
Statistical Analysis The frequency of each primary diagnosis was determined for consults seen during the 2-year study period. We then described trends in 1) the number of consults during the course of an academic year (July of a particular year through June of the next year) and 2) the number of consults seen per month between the 2 academic years studied. Statistical significance was assessed with chisquare testing. Bivariate and multivariate logistic regression was performed to determine the unadjusted and adjusted associations between patient and service specific factors, and eligibility of a given consult for billing (ie seen by the attending physician within 24 hours). For our multivariate regression the covariates included in our model consisted of age, gender, specific urology attending, urology house officer level, location (inpatient, operating room, ED) and primary diagnosis. Statistical analysis was performed using a commercial computer package (Stata®), and a 2-tailed p value of less than 0.05 was considered statistically significant.
1846
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
Table 1. Characteristics of analytic cohort 8.4 ⫾ 7.7 163 207 228 57
(25) (32) (35) (9)
213 (33) 442 (67) 115 219 304 17
(18) (33) (46) (3)
331 (51) 150 (23) 174 (27) 343 (52) 295 (45) 17 (3) 370 (56) 285 (44) 120 (18) 75 (11) 74 (11) 58 (9) 56 (9) 54 (8) 52 (8) 48 (7) 32 (5) 31 (5) 28 (4) 22 (3) 3 (0.5) 2 (0.3)
* Includes 1 consult seen by HO level II. † Includes 1 consult seen by HO level VI.
RESULTS During the study period 665 pediatric urological consults were accrued (table 1). We evaluated 0.91 consults per day (range 0 to 6), and the gender breakdown was two-thirds male and one-third female. Mean patient age was 8.4 years. Adults with pediatric urological diagnoses comprised a small percentage of consults, and one fourth of patients were younger than 12 months. Setting breakdown was split almost equally between the ED and inpatient wards, with operating room consults being relatively rare. The most common primary diagnoses were infection, obstruction/hydronephrosis and NGB (table 1). We identified 54 patients with multiple consultations, accounting for 25% of consults. The most common primary diagnoses in these recurrent cases were infection, obstruction/hydronephrosis
slope = -1.5804
70
Number of Consults
Mean ⫾ SD age (yrs) No. age (%): Younger than 1 1–10 Older than 10–18 Older than 18 No. gender (%): F M No. attending (%): 1 2 3 Other No. staffing (%): Same day Within 24 hrs Telephone only No. location (%): Emergency department Inpatient Operating room No. HO level (%): III* IV† No. primary diagnosis (%): Infection Obstruction/hydronephrosis Urinary retention/lower urinary tract symptoms/NGB Genitourinary pain Urolithiasis Foley catheter issues Postop complication Other Phimosis Hematuria Genitourinary trauma Torsion Priapism New malignancy
80 2
R = 0.1422
60 50 40 30 20 10 0 Jul
Aug
Sep
Oct
Nov
Dec Jan Month
Feb
Mar
Apr
May
Jun
Figure 1. Trend in number of consults by month during academic year.
and pain, with 50% of the cases showing a history of NGB. The trend in number of consults per month decreased slightly during the course of the academic year (fig. 1, r2 ⫽ 0.1422). The number of consults per month was lowest in February 2009 (13) and reached a peak (38) in August 2009. There was a remarkably similar trend observed in number of consults seen during each month when comparing academic years 2008 to 2009 and 2009 to 2010 (fig. 2, p ⫽ 0.82). This monthly trend persisted when stratifying the ED (p ⫽ 0.24) and inpatient (p ⫽ 0.26) consults, and comparing the 2 academic years. Overall 481 of 665 consults (72.3%) were evaluated firsthand within 24 hours by the faculty member responsible for the consultation. Unadjusted factors associated with billing eligibility included younger age (p ⬍0.001), specific attending physician (p ⫽ 0.001) and location in the inpatient wards or operating room (p ⬍0.001, table 2). According to our multivariate model, consults varied significantly based on several factors (table 3). Certain attending physicians were significantly more likely to
Figure 2. Trend in number of consults per month comparing academic years (p ⫽ 0.82) for distribution of monthly consults between academic years.
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
Table 2. Characteristics of billable vs nonbillable consults
Age (yrs): Younger than 1 1–10 Older than 10–18 Older than 18 Gender: F M Attending: 1 2 3 Other Location: Emergency department Inpatient Operating room HO level: III* IV†
No. Billable (%)
No. Nonbillable (%)
147 (90) 134 (65) 158 (69) 42 (74)
16 (10) 73 (35) 70 (31) 15 (26)
166 (78) 315 (71)
47 (22) 127 (29)
77 (67) 153 (70) 243 (80) 8 (47)
38 (33) 66 (30) 61 (20) 9 (53)
193 (56) 273 (93) 15 (88)
150 (44) 22 (7) 2 (12)
282 (76) 199 (70)
88 (24) 86 (30)
p Value ⬍0.001
0.07
0.001
⬍0.001
0.07
* Includes 1 consult seen by HO level II. † Includes 1 consult seen by HO level VI.
see consults within 24 hours, with attending 3 nearly twice as likely to bill for consults compared to attending 1. Patients seen at the operating room and the inpatient wards were significantly more likely to be seen within 24 hours compared to those seen at the ED. Finally, patients seen by more senior residents were half as likely to have the consult be billable.
DISCUSSION Before trials, protocols or initiatives, there must be descriptive data to provide the foundation for more detailed clinical research. Our study gives a benchmark for unscheduled clinical activity in pediatric urology at a tertiary referral academic medical center. This investigation also improves on similar previous reports from other specialties by initiating prospective consult tracking and standardization of data collection.8 –11 Based on tracking through the CDB, our hospital averaged nearly 1 unplanned pediatric urological consult request per day, with a maximum of 6 daily. The most common diagnoses evaluated were infection and obstruction/hydronephrosis. Assistance with management of cases of NGB and their complications was a common request, particularly among recurrent cases. In addition to assisting with acute care issues, this finding suggests opportunities to improve the chronic care of these patients. Better long-term treatment of NGB may result in fewer complications, ED visits or inpatient admissions.
1847
Several service specific factors (unrelated to patient diagnosis or acuity) predicted whether the consult had billable potential. It is noteworthy that consults involving 1 specific attending and those evaluated by the more junior house officers were more often eligible for revenue collection. The variability in staffing between the attendings reflects their different clinical and administrative duties, as well as the cooperative model for handling consult requests. The attending staffing the most consults has primarily clinical responsibilities, whereas the other 2 have more administrative responsibilities. The cooperative model, where the immediately available attending will step in and staff a consult even if not on call, makes any simple analysis of consult frequency uninformative. Approximately 28% of consults were not seen by the attending within 24 hours, which conservatively translates into $25,000 of potential uncollected revenue per year. While this figure may seem low compared to the earnings generated from procedures and scheduled clinic visits, it is the equivalent of funding the trips of 12 department members to a national meeting. Finally, a downward trend in the frequency of consultation during the academic year may represent an increasing level of experience or confidence by nonurological acute care practitioners, although further work needs to be done to verify this hypothesis. These findings highlight an important component of urological practice that has not previously been described. Pediatric urologists must be accessible Table 3. Multivariate analysis of factors associated with consult eligibility for billing Covariate Age (yrs): Younger than 1 1–10 Older than 10–18 Older than 18 Gender: F M Attending: 1 2 3 Other Location: Emergency department Inpatient Operating room HO level: III* VI†
Unadjusted OR (95% CI)
Adjusted OR (95% CI)
p Value
1.00 — 0.19 (0.11–0.36) ⬍0.001 0.25 (0.14–0.44) ⬍0.001 0.30 (0.14–0.67) 0.003
1.00 0.50 (0.24–1.05) 0.67 (0.32–1.44) 0.92 (0.36–2.37)
0.07 0.31 0.86
1.00 0.70 (0.48–1.03)
— 0.07
1.00 0.85 (0.53–1.36)
0.50
1.00 1.14 (0.71–1.86) 1.97 (1.22–3.17) 0.44 (0.16–1.23)
— 0.59 0.006 0.12
1.00 1.08 (0.61–1.92) 1.83 (1.05–3.19) 0.56 (0.17–1.80)
0.78 0.03 0.33
1.00
—
1.00
p Value
9.64 (5.94–15.65) ⬍0.001 11.24 (5.87–21.52) ⬍0.001 5.83 (1.31–25.88) 0.02 10.26 (1.58–67.47) 0.015 1.00 0.72 (0.51–1.02)
* Includes 1 consult seen by HO level II. † Includes 1 consult seen by HO level VI.
— 0.07
1.00 0.57 (0.38–0.86)
0.007
1848
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
since urgent and emergent services are required on a nearly daily basis. If this accessibility is limited, then hospital discharge could be delayed unnecessarily.12,13 One previous study demonstrated that the need for an inpatient consultation delays the time to discharge by more than 14 hours.12 In another pediatric population 14% of delayed discharges were due to a consult taking more than 24 hours, and delay in discharge at 1 facility accounted for $170,000 in excess cost during a 1-month period.13 At worst, lack of timely consultation can delay diagnosis and treatment. This factor must be considered for urology departments and groups that cover multiple sites, since the main site can be expected to generate at least 1 unscheduled bit of clinical activity daily. If staffing is planned appropriately and consults are billed appropriately, then the savings due to timely consultation could result in net savings for a health care system. Timely staffing and documentation of consults facilitate accurate billing and revenue collection. One study revealed that 20% of revenue for a pediatric surgery department was generated from nonoperative services, and the authors stressed the importance of educating faculty about the potential for lost revenue if documentation for these services is poor.7 In addition to improving billing and streamlining consult care, understanding the details of our unplanned clinical activity could result in cost savings and improved patient safety through educational initiatives. As illustrated by Thomas et al, many new interns feel uncomfortable with urethral catheterization, yet they are expected to perform this procedure with little hands-on training or supervision.5 They suggest that simple training to increase competence and confidence with this procedure could reduce the morbidity of urethral catheterization. Our study has several limitations. We relied on the resident physician entering the consult for completeness and accuracy of the data. To minimize missing data, we started data analysis 2 years after the CDB project began, once the process had become part of the regular resident routine. To minimize errors in data accuracy, we conducted a retrospective chart review on portions of the prospectively collected data. Additionally our 4 most common primary diagnoses (infection, obstruction/hydronephrosis, retention/lower urinary tract symptoms/NGB and genitourinary pain) are all often interrelated, and we rely on the judgment of the recording physi-
cian to determine the main reason a consultation is requested. However, we believe that our method of prospective data entry minimizes the possibility of incomplete and/or inaccurate data to the greatest extent feasible. Other limitations include the relatively short period of our investigation. We will continue to collect data regarding unplanned clinical activity for further evaluation of trends during a longer period. Our initial data also lack details regarding requesting service and the time of day the consultation was seen. These data could be a valuable addition to future investigations, so we added “Time of day” and “Requesting service” fields starting in July 2011. Despite these limitations, the findings from the CDB supply an important piece of information regarding the clinical activity of pediatric urologists. With the advent of the pediatric subspecialty certification from the American Board of Urology, data available from operative logs and scheduled clinical activity describe well the bulk of the scheduled pediatric urology workday.6 Our data complement and complete this description. A more complete appreciation of the workload a pediatric urologist can expect on a daily basis is now possible. Being a pediatric urologist in most active settings means one should expect to be called for something unplanned every workday. The CDB concept helps to plan efficient staffing models, develop ideas for educational initiatives, identify opportunities to improve patient care and optimize appropriate billing. These factors could ultimately enhance service and preserve resources at the hospital level. We plan to continue the CDB, with future projects including a more detailed financial analysis, description of our adult urological unplanned clinical activity and expansion of educational initiatives. Information derived from analysis of consultations to our service will improve patient care and urological education.
CONCLUSIONS We averaged nearly 1 unplanned pediatric urological consultation per day at our tertiary care academic medical center. Understanding unplanned clinical activity could help to improve patient safety, hospital efficiency and generation of departmental revenue. Tools such as the CDB could also help facilitate hypothesis generation for administrative claims database projects and inform quality improvement initiatives.
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
1849
REFERENCES 1. Brody AA, Ciemins E, Newman J et al: The effects of an inpatient palliative care team on discharge disposition. J Palliat Med 2010; 13: 541. 2. Dhillon N, Kopetz S, Pei BL et al: Clinical findings of a palliative care consultation team at a comprehensive cancer center. J Palliat Med 2008; 11: 191. 3. Mancusi S and Festa Neto C: Inpatient dermatological consultations in a university hospital. Clinics (Sao Paulo) 2010; 65: 851. 4. Penate Y, Guillermo N, Melwani P et al: Dermatologists in hospital wards: an 8-year study of dermatology consultations. Dermatology 2009; 219: 225.
5. Thomas AZ, Giri SK, Meagher D et al: Avoidable iatrogenic complications of urethral catheterization and inadequate intern training in a tertiary-care teaching hospital. BJU Int 2009; 104: 1109. 6. Kogan BA and Feustel PJ: What can we learn from pediatric urology certification logs? Urology 2011; 78: 147. 7. Gollin G and Moores D: Turning whine into wine: the fiscal impact of comprehensive documentation and billing for nonoperative pediatric surgical services. J Pediatr Surg 2006; 41: 1093. 8. Akl K: Pediatric nephrology consultations in a tertiary academic center in Jordan. Saudi J Kidney Dis Transpl 2008; 19: 456.
9. Ali E, Chaila E, Hutchinson M et al: The ‘hidden work’ of a hospital neurologist: 1000 consults later. Eur J Neurol 2010; 17: e28. 10. Geggel RL: Conditions leading to pediatric cardiology consultation in a tertiary academic hospital. Pediatrics 2004; 114: e409. 11. Penate Y, Borrego L, Hernandez N et al: Pediatric dermatology consultations: a retrospective analysis of inpatient consultations referred to the dermatology service. Pediatr Dermatol 2012; 29: 115. 12. Chen LM, Freitag MH, Franco M et al: Natural history of late discharges from a general medical ward. J Hosp Med 2009; 4: 226. 13. Srivastava R, Stone BL, Patel R et al: Delays in discharge in a tertiary care pediatric hospital. J Hosp Med 2009; 4: 481.
Abbreviations and Acronyms CDB ⫽ consult database ED ⫽ emergency department HO ⫽ house officer NGB ⫽ neurogenic bladder Submitted for publication September 16, 2011. * Correspondence: Department of Urology, Children’s Hospital Boston, 300 Longwood Ave., HU 3rd Floor, Boston, Massachusetts 02115 (telephone: 734-678-1594; FAX: 617-730-0474; e-mail: [email protected]).
Purpose: The planned clinical activity of pediatric urologists has been well described. However, little is known about nonscheduled work (eg consultation requests). We describe the unplanned clinical activity of pediatric urologists at a high volume academic medical center. Materials and Methods: Demographic data regarding inpatient, operating room and emergency department pediatric urology consults were prospectively entered into an internal database. Consults from July 2008 through June 2010 underwent retrospective chart abstraction to identify reasons for consultation. Bivariate and multivariate statistics were used to evaluate 1) temporal trends in unplanned clinical activity, and 2) patient and service specific factors associated with whether a consult was billable (ie seen by attending physician within 24 hours). Results: During the study period 665 pediatric consults were obtained. Mean ⫾ SD patient age was 8.4 ⫾ 7.7 years. Nearly all consults were seen at the emergency department (51%) or the inpatient wards (47%). The most common primary diagnoses were infection, obstruction/hydronephrosis and neurogenic bladder. The number of consults per month decreased during the course of the academic year (r2 ⫽ 0.1422). Nearly three fourths of consults were eligible for billing. The factors associated with consult eligibility for billing included specific attending physician (p ⫽ 0.03), location (p ⬍0.0001) and house officer experience (p ⫽ 0.007). Conclusions: At our academic pediatric hospital we averaged nearly 1 unplanned pediatric urology consult per day. Several service specific factors (unrelated to patient diagnosis or acuity) were associated with whether the consult had the potential to generate revenue. Unplanned clinical activity is an important factor to consider when planning departmental funding, staffing and training. Key Words: academic medical centers; pediatrics; referral and consultation; urology department, hospital
1844
www.jurology.com
PEDIATRIC urology joined the list of 147 medical specialties and subspecialties certified by the American Board of Medical Specialties in 2008. The scope of practice for each of these fields is constantly changing in response to technological, socioeconomical and regulatory forces. To finance and organize
health care delivery and education appropriately, it is essential to understand the full scope of practice and workforce expectations. Unplanned clinical consultation requests are an important component of the scope of any physician or department practice. Much of the literature
0022-5347/12/1875-1844/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION
Vol. 187, 1844-1849, May 2012 Printed in U.S.A. DOI:10.1016/j.juro.2011.12.097
AND
RESEARCH, INC.
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
regarding unplanned inpatient consultation comes from specialties such as dermatology and palliative care.1– 4 However, these specialties are consultation driven, with clinical work flow and revenue streams that differ from urological practice. The urological literature regarding unplanned clinical activity is limited but includes an investigation of consultations for iatrogenic catheter injuries.5 That study suggests that understanding the events leading to urgent urological consultation could improve patient outcomes through procedural education. In pediatric urology clinical activity has been catalogued by individual case logs, by databases of third party payers and by the American Board of Urology.6 These sources capture the activity of pediatric urologists at the administrative level but lack the granularity necessary to understand the details of hospital work flow. Therefore, little is known about the nonscheduled work done by pediatric urologists. Understanding unplanned clinical activity has the potential to inform educational initiatives (which ultimately improve patient safety), reveal patterns of documentation for appropriate billing and assist with streamlining work flow for appropriate staffing.7 Despite these potential benefits, consultations to pediatric urologists from EDs, inpatient settings and operating rooms have not previously been well described. Thus, we attempted to understand better the unplanned clinical activities in our academic pediatric urology practice.
METHODS Consult Work Flow At our institution requests for urology consultation are made by contacting the resident on call, who then evaluates the case and staffs the case with a faculty member. There is no specific pediatric urology consult service. Although on call for similar amounts of time in the evenings, HO III (junior) residents tend to be responsible for fielding consult calls during the day, which often results in the junior residents evaluating more consults than senior residents. The consult is staffed with the primary urologist of the patient (if applicable and available) or with the pediatric urologist on call. If necessary, consults are staffed with faculty members who are immediately available, even if they are not technically on call. This cooperative approach can result in different faculty members being responsible for unequal numbers of consults. Once the resident responsible for the encounter completes the documentation, details regarding the case are entered into a CDB.
Data Source An institutional CDB was created in 2006 as a means of tracking consults for quality control, education and billing. The CDB uses a password protected commercial database program (Microsoft® Access) and is housed on a secure medical center network server. Each record notes
1845
the date of service, medical identification number, patient gender, urology resident physician, urology attending physician, location (ED, operating room, inpatient, other) and type of interaction. Type of interaction is based on timing of staffing and defined as either “Seen same day” (ie same calendar day), “Seen in 24 hours” (ie not the same calendar day but within 24 hours) or “Telephone only.” The “Seen in 24 hours” category captures patients who were evaluated by the resident in the ED, then admitted and seen by the attending on the wards the next day. Cases that are managed only over the phone are listed as “Telephone only.” These options were chosen because they determined what type of billing could be made by most payers, as only consults seen within 24 hours are considered billable. Data entry was standardized by using dropdown menus for each field and entered prospectively. After an initial year to test the basic concept, we began to track fully all consults beginning in July 2007. In 2009 we added 3 fields to categorize the primary, secondary and tertiary reasons for consultation, which are determined by the physician completing the consult entry. Monthly reports of consult statistics are circulated to all resident and staff physicians. Data entry began in 2006 but the analysis for this project did not begin until 2008, when the CDB was a consistent part of work flow. Thus, we estimate the number of missing consults to be minimal.
Data Collection and Inclusion Criteria The CDB was queried for all pediatric consults seen between July 2008 and June 2010. We defined a pediatric consult as a clinical consultation for any patient younger than 18 years or a patient 18 or older still under the care of a pediatric urologist for a pediatric urological diagnosis. All consults that met this definition underwent retrospective chart review to confirm the primary, secondary and tertiary reasons for consultation. For patients seen before July 2009, when the diagnosis fields were added to the database, the reasons for consultation were obtained through chart review only.
Statistical Analysis The frequency of each primary diagnosis was determined for consults seen during the 2-year study period. We then described trends in 1) the number of consults during the course of an academic year (July of a particular year through June of the next year) and 2) the number of consults seen per month between the 2 academic years studied. Statistical significance was assessed with chisquare testing. Bivariate and multivariate logistic regression was performed to determine the unadjusted and adjusted associations between patient and service specific factors, and eligibility of a given consult for billing (ie seen by the attending physician within 24 hours). For our multivariate regression the covariates included in our model consisted of age, gender, specific urology attending, urology house officer level, location (inpatient, operating room, ED) and primary diagnosis. Statistical analysis was performed using a commercial computer package (Stata®), and a 2-tailed p value of less than 0.05 was considered statistically significant.
1846
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
Table 1. Characteristics of analytic cohort 8.4 ⫾ 7.7 163 207 228 57
(25) (32) (35) (9)
213 (33) 442 (67) 115 219 304 17
(18) (33) (46) (3)
331 (51) 150 (23) 174 (27) 343 (52) 295 (45) 17 (3) 370 (56) 285 (44) 120 (18) 75 (11) 74 (11) 58 (9) 56 (9) 54 (8) 52 (8) 48 (7) 32 (5) 31 (5) 28 (4) 22 (3) 3 (0.5) 2 (0.3)
* Includes 1 consult seen by HO level II. † Includes 1 consult seen by HO level VI.
RESULTS During the study period 665 pediatric urological consults were accrued (table 1). We evaluated 0.91 consults per day (range 0 to 6), and the gender breakdown was two-thirds male and one-third female. Mean patient age was 8.4 years. Adults with pediatric urological diagnoses comprised a small percentage of consults, and one fourth of patients were younger than 12 months. Setting breakdown was split almost equally between the ED and inpatient wards, with operating room consults being relatively rare. The most common primary diagnoses were infection, obstruction/hydronephrosis and NGB (table 1). We identified 54 patients with multiple consultations, accounting for 25% of consults. The most common primary diagnoses in these recurrent cases were infection, obstruction/hydronephrosis
slope = -1.5804
70
Number of Consults
Mean ⫾ SD age (yrs) No. age (%): Younger than 1 1–10 Older than 10–18 Older than 18 No. gender (%): F M No. attending (%): 1 2 3 Other No. staffing (%): Same day Within 24 hrs Telephone only No. location (%): Emergency department Inpatient Operating room No. HO level (%): III* IV† No. primary diagnosis (%): Infection Obstruction/hydronephrosis Urinary retention/lower urinary tract symptoms/NGB Genitourinary pain Urolithiasis Foley catheter issues Postop complication Other Phimosis Hematuria Genitourinary trauma Torsion Priapism New malignancy
80 2
R = 0.1422
60 50 40 30 20 10 0 Jul
Aug
Sep
Oct
Nov
Dec Jan Month
Feb
Mar
Apr
May
Jun
Figure 1. Trend in number of consults by month during academic year.
and pain, with 50% of the cases showing a history of NGB. The trend in number of consults per month decreased slightly during the course of the academic year (fig. 1, r2 ⫽ 0.1422). The number of consults per month was lowest in February 2009 (13) and reached a peak (38) in August 2009. There was a remarkably similar trend observed in number of consults seen during each month when comparing academic years 2008 to 2009 and 2009 to 2010 (fig. 2, p ⫽ 0.82). This monthly trend persisted when stratifying the ED (p ⫽ 0.24) and inpatient (p ⫽ 0.26) consults, and comparing the 2 academic years. Overall 481 of 665 consults (72.3%) were evaluated firsthand within 24 hours by the faculty member responsible for the consultation. Unadjusted factors associated with billing eligibility included younger age (p ⬍0.001), specific attending physician (p ⫽ 0.001) and location in the inpatient wards or operating room (p ⬍0.001, table 2). According to our multivariate model, consults varied significantly based on several factors (table 3). Certain attending physicians were significantly more likely to
Figure 2. Trend in number of consults per month comparing academic years (p ⫽ 0.82) for distribution of monthly consults between academic years.
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
Table 2. Characteristics of billable vs nonbillable consults
Age (yrs): Younger than 1 1–10 Older than 10–18 Older than 18 Gender: F M Attending: 1 2 3 Other Location: Emergency department Inpatient Operating room HO level: III* IV†
No. Billable (%)
No. Nonbillable (%)
147 (90) 134 (65) 158 (69) 42 (74)
16 (10) 73 (35) 70 (31) 15 (26)
166 (78) 315 (71)
47 (22) 127 (29)
77 (67) 153 (70) 243 (80) 8 (47)
38 (33) 66 (30) 61 (20) 9 (53)
193 (56) 273 (93) 15 (88)
150 (44) 22 (7) 2 (12)
282 (76) 199 (70)
88 (24) 86 (30)
p Value ⬍0.001
0.07
0.001
⬍0.001
0.07
* Includes 1 consult seen by HO level II. † Includes 1 consult seen by HO level VI.
see consults within 24 hours, with attending 3 nearly twice as likely to bill for consults compared to attending 1. Patients seen at the operating room and the inpatient wards were significantly more likely to be seen within 24 hours compared to those seen at the ED. Finally, patients seen by more senior residents were half as likely to have the consult be billable.
DISCUSSION Before trials, protocols or initiatives, there must be descriptive data to provide the foundation for more detailed clinical research. Our study gives a benchmark for unscheduled clinical activity in pediatric urology at a tertiary referral academic medical center. This investigation also improves on similar previous reports from other specialties by initiating prospective consult tracking and standardization of data collection.8 –11 Based on tracking through the CDB, our hospital averaged nearly 1 unplanned pediatric urological consult request per day, with a maximum of 6 daily. The most common diagnoses evaluated were infection and obstruction/hydronephrosis. Assistance with management of cases of NGB and their complications was a common request, particularly among recurrent cases. In addition to assisting with acute care issues, this finding suggests opportunities to improve the chronic care of these patients. Better long-term treatment of NGB may result in fewer complications, ED visits or inpatient admissions.
1847
Several service specific factors (unrelated to patient diagnosis or acuity) predicted whether the consult had billable potential. It is noteworthy that consults involving 1 specific attending and those evaluated by the more junior house officers were more often eligible for revenue collection. The variability in staffing between the attendings reflects their different clinical and administrative duties, as well as the cooperative model for handling consult requests. The attending staffing the most consults has primarily clinical responsibilities, whereas the other 2 have more administrative responsibilities. The cooperative model, where the immediately available attending will step in and staff a consult even if not on call, makes any simple analysis of consult frequency uninformative. Approximately 28% of consults were not seen by the attending within 24 hours, which conservatively translates into $25,000 of potential uncollected revenue per year. While this figure may seem low compared to the earnings generated from procedures and scheduled clinic visits, it is the equivalent of funding the trips of 12 department members to a national meeting. Finally, a downward trend in the frequency of consultation during the academic year may represent an increasing level of experience or confidence by nonurological acute care practitioners, although further work needs to be done to verify this hypothesis. These findings highlight an important component of urological practice that has not previously been described. Pediatric urologists must be accessible Table 3. Multivariate analysis of factors associated with consult eligibility for billing Covariate Age (yrs): Younger than 1 1–10 Older than 10–18 Older than 18 Gender: F M Attending: 1 2 3 Other Location: Emergency department Inpatient Operating room HO level: III* VI†
Unadjusted OR (95% CI)
Adjusted OR (95% CI)
p Value
1.00 — 0.19 (0.11–0.36) ⬍0.001 0.25 (0.14–0.44) ⬍0.001 0.30 (0.14–0.67) 0.003
1.00 0.50 (0.24–1.05) 0.67 (0.32–1.44) 0.92 (0.36–2.37)
0.07 0.31 0.86
1.00 0.70 (0.48–1.03)
— 0.07
1.00 0.85 (0.53–1.36)
0.50
1.00 1.14 (0.71–1.86) 1.97 (1.22–3.17) 0.44 (0.16–1.23)
— 0.59 0.006 0.12
1.00 1.08 (0.61–1.92) 1.83 (1.05–3.19) 0.56 (0.17–1.80)
0.78 0.03 0.33
1.00
—
1.00
p Value
9.64 (5.94–15.65) ⬍0.001 11.24 (5.87–21.52) ⬍0.001 5.83 (1.31–25.88) 0.02 10.26 (1.58–67.47) 0.015 1.00 0.72 (0.51–1.02)
* Includes 1 consult seen by HO level II. † Includes 1 consult seen by HO level VI.
— 0.07
1.00 0.57 (0.38–0.86)
0.007
1848
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
since urgent and emergent services are required on a nearly daily basis. If this accessibility is limited, then hospital discharge could be delayed unnecessarily.12,13 One previous study demonstrated that the need for an inpatient consultation delays the time to discharge by more than 14 hours.12 In another pediatric population 14% of delayed discharges were due to a consult taking more than 24 hours, and delay in discharge at 1 facility accounted for $170,000 in excess cost during a 1-month period.13 At worst, lack of timely consultation can delay diagnosis and treatment. This factor must be considered for urology departments and groups that cover multiple sites, since the main site can be expected to generate at least 1 unscheduled bit of clinical activity daily. If staffing is planned appropriately and consults are billed appropriately, then the savings due to timely consultation could result in net savings for a health care system. Timely staffing and documentation of consults facilitate accurate billing and revenue collection. One study revealed that 20% of revenue for a pediatric surgery department was generated from nonoperative services, and the authors stressed the importance of educating faculty about the potential for lost revenue if documentation for these services is poor.7 In addition to improving billing and streamlining consult care, understanding the details of our unplanned clinical activity could result in cost savings and improved patient safety through educational initiatives. As illustrated by Thomas et al, many new interns feel uncomfortable with urethral catheterization, yet they are expected to perform this procedure with little hands-on training or supervision.5 They suggest that simple training to increase competence and confidence with this procedure could reduce the morbidity of urethral catheterization. Our study has several limitations. We relied on the resident physician entering the consult for completeness and accuracy of the data. To minimize missing data, we started data analysis 2 years after the CDB project began, once the process had become part of the regular resident routine. To minimize errors in data accuracy, we conducted a retrospective chart review on portions of the prospectively collected data. Additionally our 4 most common primary diagnoses (infection, obstruction/hydronephrosis, retention/lower urinary tract symptoms/NGB and genitourinary pain) are all often interrelated, and we rely on the judgment of the recording physi-
cian to determine the main reason a consultation is requested. However, we believe that our method of prospective data entry minimizes the possibility of incomplete and/or inaccurate data to the greatest extent feasible. Other limitations include the relatively short period of our investigation. We will continue to collect data regarding unplanned clinical activity for further evaluation of trends during a longer period. Our initial data also lack details regarding requesting service and the time of day the consultation was seen. These data could be a valuable addition to future investigations, so we added “Time of day” and “Requesting service” fields starting in July 2011. Despite these limitations, the findings from the CDB supply an important piece of information regarding the clinical activity of pediatric urologists. With the advent of the pediatric subspecialty certification from the American Board of Urology, data available from operative logs and scheduled clinical activity describe well the bulk of the scheduled pediatric urology workday.6 Our data complement and complete this description. A more complete appreciation of the workload a pediatric urologist can expect on a daily basis is now possible. Being a pediatric urologist in most active settings means one should expect to be called for something unplanned every workday. The CDB concept helps to plan efficient staffing models, develop ideas for educational initiatives, identify opportunities to improve patient care and optimize appropriate billing. These factors could ultimately enhance service and preserve resources at the hospital level. We plan to continue the CDB, with future projects including a more detailed financial analysis, description of our adult urological unplanned clinical activity and expansion of educational initiatives. Information derived from analysis of consultations to our service will improve patient care and urological education.
CONCLUSIONS We averaged nearly 1 unplanned pediatric urological consultation per day at our tertiary care academic medical center. Understanding unplanned clinical activity could help to improve patient safety, hospital efficiency and generation of departmental revenue. Tools such as the CDB could also help facilitate hypothesis generation for administrative claims database projects and inform quality improvement initiatives.
PROSPECTIVE TRACKING OF PEDIATRIC UROLOGY CONSULTS
1849
REFERENCES 1. Brody AA, Ciemins E, Newman J et al: The effects of an inpatient palliative care team on discharge disposition. J Palliat Med 2010; 13: 541. 2. Dhillon N, Kopetz S, Pei BL et al: Clinical findings of a palliative care consultation team at a comprehensive cancer center. J Palliat Med 2008; 11: 191. 3. Mancusi S and Festa Neto C: Inpatient dermatological consultations in a university hospital. Clinics (Sao Paulo) 2010; 65: 851. 4. Penate Y, Guillermo N, Melwani P et al: Dermatologists in hospital wards: an 8-year study of dermatology consultations. Dermatology 2009; 219: 225.
5. Thomas AZ, Giri SK, Meagher D et al: Avoidable iatrogenic complications of urethral catheterization and inadequate intern training in a tertiary-care teaching hospital. BJU Int 2009; 104: 1109. 6. Kogan BA and Feustel PJ: What can we learn from pediatric urology certification logs? Urology 2011; 78: 147. 7. Gollin G and Moores D: Turning whine into wine: the fiscal impact of comprehensive documentation and billing for nonoperative pediatric surgical services. J Pediatr Surg 2006; 41: 1093. 8. Akl K: Pediatric nephrology consultations in a tertiary academic center in Jordan. Saudi J Kidney Dis Transpl 2008; 19: 456.
9. Ali E, Chaila E, Hutchinson M et al: The ‘hidden work’ of a hospital neurologist: 1000 consults later. Eur J Neurol 2010; 17: e28. 10. Geggel RL: Conditions leading to pediatric cardiology consultation in a tertiary academic hospital. Pediatrics 2004; 114: e409. 11. Penate Y, Borrego L, Hernandez N et al: Pediatric dermatology consultations: a retrospective analysis of inpatient consultations referred to the dermatology service. Pediatr Dermatol 2012; 29: 115. 12. Chen LM, Freitag MH, Franco M et al: Natural history of late discharges from a general medical ward. J Hosp Med 2009; 4: 226. 13. Srivastava R, Stone BL, Patel R et al: Delays in discharge in a tertiary care pediatric hospital. J Hosp Med 2009; 4: 481.