- Email: [email protected]
Effect of season, age, and gender on renal colic incidence
Effect of Season, Age, and Gender on Renal Colic Incidence VEENA CHAUHAN, BARNET ESKIN, MD, PHD, JOHN R. ALLEGRA, MD, PHD, AND DENNIS G. COCHRANE, MD Our objective was to examine the effect of ambient temperature, age, and gender on the incidence of emergency department (ED) renal colic visits. We retrospectively analyzed a database of 15 New Jersey EDs from January 1, 1996 to December 31, 2002. We analyzed the number of renal colic visits as a fraction of total visits in monthly intervals. We used the Chi-squared test and Pearson’s correlation coefficient, with P<.05 taken as statistically significant. Of the 3.5 million patient visits in the database, 30,358 (0.9%) had renal colic. Renal colic visits were 16% more likely in warmer than colder months (P<.001) and this effect was greatest in older patients and males. We conclude that higher ambient temperature, older age and male gender are associated with increased incidence of ED renal colic visits. Advice to patients, especially older males, to avoid dehydration particularly during hot weather may help prevent bouts of renal colic. (Am J Emerg Med 2004;22:560-563. © 2004 Elsevier Inc. All rights reserved.)
Although several studies, using medical records and hospital admissions and discharges, found no seasonal variation in urinary calculi formation,1-3 others did find seasonal variation.4-9 In the most recent study, in Saudi Arabia, using a database of 80,951 ED visits, of which 447 were for renal colic, the incidence of renal colic visits was significantly higher in the warmer months of June, July, and August.9 The objective of our study was to determine if renal colic visits in northern New Jersey exhibited seasonal patterns, using a database that was much larger than that available in previous studies. We also wished to examine the effect of age and gender. We hypothesized that renal colic visits would increase more in older patients in the warmer months and that there would be no gender differences. METHODS We retrospectively analyzed computerized billing records for patient visits at 15 northern New Jersey EDs from January 1, 1996 to December 31, 2002. Consecutive patients seen by ED physicians were included. Emergency
TABLE 1.
ICD-9 Codes Used
Calculus kidney Calculus ureter Urinary calculus, unspecified Bladder calculus Urethral calculus Renal colic
592.0 592.1 592.9 594.1 594.2 788.0
physicians see 80%-90% of all patient visits at these EDs. The EDs are in urban and suburban areas and in both teaching and nonteaching hospitals. Total ED annual volumes range from 20,000 to 70,000. The physicians’ charts are coded by trained coders according to the International Classification of Diseases 9th revision and Clinical Modification (ICD-9) codes. Patients were considered to have renal colic if the primary, secondary or tertiary diagnosis included any of the ICD-9 codes listed in Table 1. We analyzed the number of renal colic visits as a fraction of total visits in monthly intervals. Months were then separated into 3-month periods, with June, July, and August considered “warmer months” and December, January, and February, “colder months.” We correlated renal colic visits with mean monthly ambient temperatures in Northern NJ10 using Pearson’s correlation coefficients. To examine age differences we a priori divided the renal colic patients into age quartiles. We calculated the ratio of the incidence of renal colic visits in the “warmer” to the “colder” months and calculated this ratio for each age quartile for all patients and for males and females separately. We used the Chisquared test, with P⬍.05 taken as statistically significant. All analysis was conducted using MS Excel 97 (Microsoft, Redmond, WA) and SPSS6 (SPSS, Cary, NC), . The Institutional Review Board at the authors’ institution approved the study. RESULTS
From the Morristown Memorial Hospital Residency in Emergency Medicine, Morristown, NJ. Manuscript received October 26, 2003; accepted October 26, 2003. Supported in part by the Emergency Medical Associates Research Foundation. Address reprint requests to Barnet Eskin, MD, PhD, Department Of Emergency Medicine, Morristown Memorial Hospital, 100 Madison Avenue, Morristown, NJ. Email: [email protected] Key Words: renal colic, incidence, emergency departments, epidemiology © 2004 Elsevier Inc. All rights reserved. 0735-6757/04/2207-0011$30.00/0 doi:10.1016/j.ajem.2004.08.016 560
There were 3.5 million patient visits in the database; of which 30,358 (0.9%) were for renal colic. The renal colic patients had a mean age of 44 ⫹/- 14 years (Fig 1) and 63% were males. We found significant seasonal variations in renal colic visits. The renal colic visits as a fraction of total ED visits grouped by three-month periods are shown in Fig 2. We found that renal colic visits were 16% more likely in the warmer months of June, July, and August compared with the colder months of December, January, and February (P⬍.001).
CHAUHAN ET AL ■ RENAL COLIC INCIDENCE AND SEASON, AGE AND GENDER
561
FIGURE 1. Age distribution of renal colic patients.
Figure 3 shows the correlation between monthly mean ambient temperature and renal colic visit rates. There is a positive correlation between ambient temperature and renal colic visit rates, with R⫽0.42 (95% CI, 0.23 to 0.59, P⬍.001). An increase in ambient temperature from 0°C to 20°C would be expected to result in a 14% increase in the rate of renal colic visits. Figure 4 shows the ratio of renal colic visits in the warmer and colder months for each age quartile. There was a greater seasonal difference for the older age groups. The oldest age quartile has a 20% greater warmer to colder ratio of renal colic visits compared with the youngest age quartile (P⬍.001). There was a 26% greater warmer months to colder months ratio in males compared with females (P⬍.001). There is a greater change in the warmer months to colder months ratio as a function of age in males compared with females (P⬍.001, Fig 5).
DISCUSSION Our data shows a clear seasonal variation of renal colic, with a 16% increased incidence of renal colic visits in the warmer months of June, July, and August compared with the colder months of December, January, and February. This confirms the trend found in the Saudi Arabia study. We had the advantage of a larger database, however, with 68 times as many renal colic visits as reported in the Saudi Arabia study. We also found a strong positive correlation between ambient temperature and renal colic visits. An increase in ambient temperature from 0° to 20° would be expected to result in a 14% increase in the proportion of renal colic visits. The Saudi Arabia study was done in males only and so could not examine the effects of gender on the incidence of renal colic. In addition, it did not report any effects of age either. We found that the oldest age quartile has a 20% greater warmer to colder ratio of renal colic visits compared
FIGURE 2. Variation of the number of renal colic visits as a fraction of total ED visits in 3 monthly periods.
562
AMERICAN JOURNAL OF EMERGENCY MEDICINE ■ Volume 22, Number 7 ■ November 2004
FIGURE 3. Correlation between the number of renal colic visits as a fraction of total ED visits and mean monthly ambient temperature in degrees Celsius.
with the youngest age quartile. The magnitude of this effect was a 26% greater warmer to colder ratio in males compared with females. The seasonal pattern was more pronounced in the oldest age quartile and in male patients. This suggests that renal colic in older men is more heavily influenced by the hot weather and may be due to dehydration. Therefore this higher risk population should be made aware of the importance of maintaining good hydration, especially during the warmer months. LIMITATIONS There are several limitations to our study. Our findings may be influenced by shifts in population due to vaca-
FIGURE 4. Summer to winter ratio of renal colic visits for age quartiles.
tions or living in other areas during parts of the year. There may be a delay between renal stone formation and clinical manifestation. Another limitation arises from the use of ICD-9 codes. These codes are assigned based on the discharge diagnosis of the ED physician and there may be some variability in how physicians identify renal colic patients. Since this study included only those renal colic patients seen in the ED, patients evaluated in non-ED locations were excluded, as were patients seen in the ED by their private physicians. However, none of the above factors should bias the monthly patterns seen. The data here may not be representative of all hospital EDs, particularly inner city and rural hospital EDs. The sea-
CHAUHAN ET AL ■ RENAL COLIC INCIDENCE AND SEASON, AGE AND GENDER
563
FIGURE 5. Age and gender patterns of renal colic visits: summer to winter ratio.
sonal patterns reflect the climate of northern New Jersey and many not apply to other regions. CONCLUSION Our data shows there is a seasonal pattern in renal colic visits, with more renal colic visits in the warmer months. This effect is greatest in the elderly and in males Advice to patients, especially men over age 55 to avoid dehydration particularly during hot weather may help prevent bouts of renal colic. REFERENCES 1. Almby B, Meirik O, Schonebeck J: Incidence, morbidity and complications of renal and urethral calculi in a well defined geographical area. Scand J Urol Nephrol 1975;9:249-53 2. Juuti M, Heinomen OP: Incidence of uroliathiasis leading to hospitalization in Finland. Acta Med Scand 1979;206:397-403
3. Ahlstrand C, Tiselius HG: Renal stone disease in a Swedish district during one year. Scand J Urol Nephrol 1981;15:143-613 4. Robertson WG, Peacock M, Marshall RW, et al: Seasonal variations in the composition of urine in relation to calcium stoneformation. Clin Sci Mol Med 1975;49:597-602 5. Al-Dabbagh TQ, Fahadi K: Seasonal variation in the incidence of ureteric colic. Br J Urol 1977;49:269-75 6. Fujitika K: Weather and the incidence of urinary stone colic. J Urol 1979;121:318-9 7. Hussain E, Biliimoria FR, Singh PP: Urolithiasis in northeast Bombay: seasonal prevalence and chemical composition of stones. Int Urol Nephrol 1990;22:119-24 8. Baker PW, Coyle P, Bais R, et al: Influence of season, age, and sex on renal stone formation in South Australia. Med J Aust 1993; 159:390-2 9. Al-Hadramy MS: Seasonal variations of urinary stone colic in Arabia. J Pak Med Assoc 1997;47:281-4 10. Office of the NJ State Climatologist, Rutgers University: Monthly mean temperatures in Northern New Jersey (Division 1) from 1895-2003. http://climate.rutgers.edu/stateclim/data/north_ njhisttemp.html Viewed 05/12/03
Although several studies, using medical records and hospital admissions and discharges, found no seasonal variation in urinary calculi formation,1-3 others did find seasonal variation.4-9 In the most recent study, in Saudi Arabia, using a database of 80,951 ED visits, of which 447 were for renal colic, the incidence of renal colic visits was significantly higher in the warmer months of June, July, and August.9 The objective of our study was to determine if renal colic visits in northern New Jersey exhibited seasonal patterns, using a database that was much larger than that available in previous studies. We also wished to examine the effect of age and gender. We hypothesized that renal colic visits would increase more in older patients in the warmer months and that there would be no gender differences. METHODS We retrospectively analyzed computerized billing records for patient visits at 15 northern New Jersey EDs from January 1, 1996 to December 31, 2002. Consecutive patients seen by ED physicians were included. Emergency
TABLE 1.
ICD-9 Codes Used
Calculus kidney Calculus ureter Urinary calculus, unspecified Bladder calculus Urethral calculus Renal colic
592.0 592.1 592.9 594.1 594.2 788.0
physicians see 80%-90% of all patient visits at these EDs. The EDs are in urban and suburban areas and in both teaching and nonteaching hospitals. Total ED annual volumes range from 20,000 to 70,000. The physicians’ charts are coded by trained coders according to the International Classification of Diseases 9th revision and Clinical Modification (ICD-9) codes. Patients were considered to have renal colic if the primary, secondary or tertiary diagnosis included any of the ICD-9 codes listed in Table 1. We analyzed the number of renal colic visits as a fraction of total visits in monthly intervals. Months were then separated into 3-month periods, with June, July, and August considered “warmer months” and December, January, and February, “colder months.” We correlated renal colic visits with mean monthly ambient temperatures in Northern NJ10 using Pearson’s correlation coefficients. To examine age differences we a priori divided the renal colic patients into age quartiles. We calculated the ratio of the incidence of renal colic visits in the “warmer” to the “colder” months and calculated this ratio for each age quartile for all patients and for males and females separately. We used the Chisquared test, with P⬍.05 taken as statistically significant. All analysis was conducted using MS Excel 97 (Microsoft, Redmond, WA) and SPSS6 (SPSS, Cary, NC), . The Institutional Review Board at the authors’ institution approved the study. RESULTS
From the Morristown Memorial Hospital Residency in Emergency Medicine, Morristown, NJ. Manuscript received October 26, 2003; accepted October 26, 2003. Supported in part by the Emergency Medical Associates Research Foundation. Address reprint requests to Barnet Eskin, MD, PhD, Department Of Emergency Medicine, Morristown Memorial Hospital, 100 Madison Avenue, Morristown, NJ. Email: [email protected] Key Words: renal colic, incidence, emergency departments, epidemiology © 2004 Elsevier Inc. All rights reserved. 0735-6757/04/2207-0011$30.00/0 doi:10.1016/j.ajem.2004.08.016 560
There were 3.5 million patient visits in the database; of which 30,358 (0.9%) were for renal colic. The renal colic patients had a mean age of 44 ⫹/- 14 years (Fig 1) and 63% were males. We found significant seasonal variations in renal colic visits. The renal colic visits as a fraction of total ED visits grouped by three-month periods are shown in Fig 2. We found that renal colic visits were 16% more likely in the warmer months of June, July, and August compared with the colder months of December, January, and February (P⬍.001).
CHAUHAN ET AL ■ RENAL COLIC INCIDENCE AND SEASON, AGE AND GENDER
561
FIGURE 1. Age distribution of renal colic patients.
Figure 3 shows the correlation between monthly mean ambient temperature and renal colic visit rates. There is a positive correlation between ambient temperature and renal colic visit rates, with R⫽0.42 (95% CI, 0.23 to 0.59, P⬍.001). An increase in ambient temperature from 0°C to 20°C would be expected to result in a 14% increase in the rate of renal colic visits. Figure 4 shows the ratio of renal colic visits in the warmer and colder months for each age quartile. There was a greater seasonal difference for the older age groups. The oldest age quartile has a 20% greater warmer to colder ratio of renal colic visits compared with the youngest age quartile (P⬍.001). There was a 26% greater warmer months to colder months ratio in males compared with females (P⬍.001). There is a greater change in the warmer months to colder months ratio as a function of age in males compared with females (P⬍.001, Fig 5).
DISCUSSION Our data shows a clear seasonal variation of renal colic, with a 16% increased incidence of renal colic visits in the warmer months of June, July, and August compared with the colder months of December, January, and February. This confirms the trend found in the Saudi Arabia study. We had the advantage of a larger database, however, with 68 times as many renal colic visits as reported in the Saudi Arabia study. We also found a strong positive correlation between ambient temperature and renal colic visits. An increase in ambient temperature from 0° to 20° would be expected to result in a 14% increase in the proportion of renal colic visits. The Saudi Arabia study was done in males only and so could not examine the effects of gender on the incidence of renal colic. In addition, it did not report any effects of age either. We found that the oldest age quartile has a 20% greater warmer to colder ratio of renal colic visits compared
FIGURE 2. Variation of the number of renal colic visits as a fraction of total ED visits in 3 monthly periods.
562
AMERICAN JOURNAL OF EMERGENCY MEDICINE ■ Volume 22, Number 7 ■ November 2004
FIGURE 3. Correlation between the number of renal colic visits as a fraction of total ED visits and mean monthly ambient temperature in degrees Celsius.
with the youngest age quartile. The magnitude of this effect was a 26% greater warmer to colder ratio in males compared with females. The seasonal pattern was more pronounced in the oldest age quartile and in male patients. This suggests that renal colic in older men is more heavily influenced by the hot weather and may be due to dehydration. Therefore this higher risk population should be made aware of the importance of maintaining good hydration, especially during the warmer months. LIMITATIONS There are several limitations to our study. Our findings may be influenced by shifts in population due to vaca-
FIGURE 4. Summer to winter ratio of renal colic visits for age quartiles.
tions or living in other areas during parts of the year. There may be a delay between renal stone formation and clinical manifestation. Another limitation arises from the use of ICD-9 codes. These codes are assigned based on the discharge diagnosis of the ED physician and there may be some variability in how physicians identify renal colic patients. Since this study included only those renal colic patients seen in the ED, patients evaluated in non-ED locations were excluded, as were patients seen in the ED by their private physicians. However, none of the above factors should bias the monthly patterns seen. The data here may not be representative of all hospital EDs, particularly inner city and rural hospital EDs. The sea-
CHAUHAN ET AL ■ RENAL COLIC INCIDENCE AND SEASON, AGE AND GENDER
563
FIGURE 5. Age and gender patterns of renal colic visits: summer to winter ratio.
sonal patterns reflect the climate of northern New Jersey and many not apply to other regions. CONCLUSION Our data shows there is a seasonal pattern in renal colic visits, with more renal colic visits in the warmer months. This effect is greatest in the elderly and in males Advice to patients, especially men over age 55 to avoid dehydration particularly during hot weather may help prevent bouts of renal colic. REFERENCES 1. Almby B, Meirik O, Schonebeck J: Incidence, morbidity and complications of renal and urethral calculi in a well defined geographical area. Scand J Urol Nephrol 1975;9:249-53 2. Juuti M, Heinomen OP: Incidence of uroliathiasis leading to hospitalization in Finland. Acta Med Scand 1979;206:397-403
3. Ahlstrand C, Tiselius HG: Renal stone disease in a Swedish district during one year. Scand J Urol Nephrol 1981;15:143-613 4. Robertson WG, Peacock M, Marshall RW, et al: Seasonal variations in the composition of urine in relation to calcium stoneformation. Clin Sci Mol Med 1975;49:597-602 5. Al-Dabbagh TQ, Fahadi K: Seasonal variation in the incidence of ureteric colic. Br J Urol 1977;49:269-75 6. Fujitika K: Weather and the incidence of urinary stone colic. J Urol 1979;121:318-9 7. Hussain E, Biliimoria FR, Singh PP: Urolithiasis in northeast Bombay: seasonal prevalence and chemical composition of stones. Int Urol Nephrol 1990;22:119-24 8. Baker PW, Coyle P, Bais R, et al: Influence of season, age, and sex on renal stone formation in South Australia. Med J Aust 1993; 159:390-2 9. Al-Hadramy MS: Seasonal variations of urinary stone colic in Arabia. J Pak Med Assoc 1997;47:281-4 10. Office of the NJ State Climatologist, Rutgers University: Monthly mean temperatures in Northern New Jersey (Division 1) from 1895-2003. http://climate.rutgers.edu/stateclim/data/north_ njhisttemp.html Viewed 05/12/03