Article Figures & Data
Tables
Characteristic Total No Opioids Any Opioids P Value No. (%) 53,982 (100) 45,449 (84.2) 8,533 (15.8) Age, y, mean (SD) 45.6 (17.5) 45.2 (17.5) 47.4 (17.2) <.001 Race, no. (%) White 24,732 (45.8) 20,102 (44.2) 4,630 (54.3) <.001 Hispanic 13,779 (25.5) 12,297 (27.1) 1,482 (17.4) Black 10,774 (20.0) 8,824 (19.4) 1,950 (22.9) Other 4,697 (8.7) 4,226 (9.3) 471 (5.5) Education, no. (%) Less than high school 4,717 (8.7) 4,102 (9.0) 615 (7.2) <.001 Some high school 7,015 (13.0) 5,744 (12.6) 1,271 (14.9) High school graduate 16,349 (30.3) 13,577 (29.9) 2,772 (32.5) Some college 13,721 (25.4) 11,337 (24.9) 2,384 (27.9) College graduate 12,180 (22.6) 10,689 (23.5) 1,491 (17.5) Household income, % federal poverty level, no. (%) <100% 11,274 (20.9) 9,059 (19.9) 2,215 (26.0) <.001 100%-124% 3,505 (6.5) 2,895 (6.4) 610 (7.1) 125%-199% 9,102 (16.9) 7,643 (16.8) 1,459 (17.1) 200%-399% 15,494 (28.7) 13,175 (29.0) 2,319 (27.2) ≥400% 14,607 (27.1) 12,677 (27.9) 1,930 (22.6) Census region, no. (%) Northeast 8,390 (15.5) 7,346 (16.2) 1,044 (12.2) <.001 Midwest 10,646 (19.7) 8,695 (19.1) 1,951 (22.9) South 20,811 (38.6) 17,288 (38.0) 3,523 (41.3) West 14,135 (26.2) 12,120 (26.7) 2,015 (23.6) Insurance status, no. (%) Private 31,131 (57.7) 26,440 (58.2) 4,691 (55.0) <.001 Public 13,695 (25.4) 10,624 (23.4) 3,071 (36.0) None 9,156 (17.0) 8,385 (18.4) 771 (9.0) 12-Item Short-Form Health Survey component, mean (SD) Physical Component Summary score 48.7 (11.1) 50.0 (10.0) 41.6 (13.4) <.001 Mental Component Summary score 49.9 (10.4) 50.4 (10.0) 46.7 (11.9) <.001 Self-rated health, no. (%) Excellent 11,481 (21.3) 10,508 (23.1) 973 (11.4) <.001 Very good 17,285 (32.0) 15,265 (33.6) 2,020 (23.7) Good 16,309 (30.2) 13,541 (29.8) 2,768 (32.4) Fair 6,948 (12.9) 5,079 (11.2) 1,869 (21.9) Poor 1,959 (3.6) 1,056 (2.3) 903 (10.6) Current smoker, no. (%) No 45,543 (84.4) 39,127 (86.1) 6,416 (75.2) <.001 Yes 8,439 (15.6) 6,322 (13.9) 2,117 (24.8) Count of chronic diseases, 0 0 1.0 <.001 median (IQR) (0-1.0) (0-1.0) (0-2.0) Usual source of care, no. (%) No 12,534 (23.2) 11,337 (24.9) 1,197 (14.0) <.001 Yes 41,448 (76.8) 34,112 (75.1) 7,336 (86.0) Number of office visits, 2.0 1.0 5.0 <.001 median (IQR) (0-5.0) (0-4.0) (2.0-10.0) Opioid category, no. (%) 0 prescriptions 45,449 (84.2) 45,449 (100.0) <.001 1 prescription 4,505 (8.3) 4,505 (52.8) 2-3 prescriptions 1,779 (3.3) 1,779 (20.8) 4-11 prescriptions 1,622 (3.0) 1,622 (19.0) >12 prescriptions 627 (1.2) 627 (7.3) Cervical cancer screening, no. (%) Yes 39,229 (84.2) 32,914 (84.1) 6,315 (84.8) .13 Breast cancer screening, no. (%) Yes 23,449 (70.7) 19,220 (70.2) 4,229 (73.2) <.001 Colorectal cancer screening, no. (%) Yes 11,622 (49.2) 9,276 (47.8) 2,346 (55.8) <.001 IQR = interquartile range.
Notes: Data not adjusted for survey characteristics. P values based on sample and not survey adjusted.
Characteristic AOR (95% CI) P Value Age 0.98 (0.98-0.98) <.001 Race/ethnicity (ref = white) Hispanic 0.67 (0.61-0.73) <.001 Black 0.91 (0.84-0.99) .023 Other 0.6 (0.52-0.69) <.001 Education (ref = less than high school) Some high school 1.18 (1.02-1.36) .028 High school graduate 1.21 (1.05-1.39) .008 Some college 1.29 (1.12-1.47) <.001 College graduate 1.08 (0.93-1.25) .299 Household income, % federal poverty level (ref = <100%) 100%-124% 0.97 (0.84-1.11) .643 125%-199% 0.96 (0.87-1.05) .332 200%-399% 0.88 (0.8-0.97) .014 ≥400% 0.84 (0.75-0.94) .003 Census region (ref = Northeast) Midwest 1.59 (1.43-1.77) <.001 South 1.58 (1.42-1.76) <.001 West 1.7 (1.52-1.9) <.001 Insurance status (ref = private) Public 0.95 (0.87-1.03) .206 None 0.59 (0.53-0.67) <.001 12-Item Short-Form Health Survey component Physical Component Summary score 0.95 (0.95-0.96) <.001 Mental Component Summary score 0.99 (0.98-0.99) <.001 Self-rated health 1.07 (1.03-1.11) .001 Current smoker (vs nonsmoker) 1.56 (1.45-1.67) <.001 Count of chronic diseases 1.05 (1.02-1.09) .001 Usual source of care, Yes (vs No) 1.43 (1.31-1.56) <.001 Number of office visits 1.08 (1.07-1.08) <.001 AOR = adjusted odds ratio.
Note: Data adjusted for survey characteristics and panel year.
- Table 3
Adjusted Odds of Cancer Screening, Without and With Health Care Utilization Adjustment
Screening Test Model 1: Without Utilization Adjustment, AOR (95% CI) Model 2: With Utilization Adjustment, AOR (95% CI) Statistical Significance of Difference Between Models (P Value)a Breast cancer (n = 33,166) 1.26 (1.16-1.38) P<.001 1.07 (0.98-1.18) P = .09 <.001 Cervical cancer (n = 46,598) 1.22 (1.13-1.33) P<.001 1.01 (0.93-1.09) P = .8 <.001 Colorectal cancer (n = 23,613) 1.22 (1.12-1.33) P<.001 1.04 (0.95-1.14) P = .35 <.001 AOR = adjusted odds ratio for women reporting opioid prescriptions vs women reporting none.
Notes: Model 1 includes adjustment for age, race/ethnicity, US Census region, education level, income, health insurance status, health status, comorbidities, smoking status, usual source of care, and panel year. Model 2 includes all covariates in Model 1, as well as count of doctor’s office visits.
↵a Adjusted Wald test comparing parameter estimates for opioids in Model 1 and Model 2.
Additional Files
The Article in Brief
Cancer Screening Among Women Prescribed Opioids: A National Study
Alicia Agnoli , and colleagues
Background Concerns have been raised that in the primary care setting, treating chronic pain and managing opioid prescriptions may be associated with negative preventive care outcomes. Managing patient pain and prescription opioids takes considerable time, and these competing demands may strain and impair the delivery of evidence-based preventive health needs, such as recommended cancer screenings.
What This Study Found Researchers at the University of California, Davis analyzed data from a nationally representative sample of 53,982 women in the United States. Findings revealed that women who are prescribed opioids were more likely to receive breast, cervical, and colorectal cancer screenings for the simple fact that they are frequent users of the health care system. They had a median number of doctor visits that was five times higher than their non-prescribed counterparts. When this factor was controlled for, analysis showed no association between prescription opioid use and cancer screening. Authors conclude that U.S. women who take prescription opioids are no less likely to receive key cancer screenings when compared to women who are not prescribed opioids.
Implications
- This study is one of the first to examine access to key preventive health services for opioid versus non-opioid users. Authors suggest that "the key driver of whether women receive recommended cancer screening is simply how often they see the doctor."