Table 1. Effectiveness of a Mailed Human Papillomavirus Kit Intervention vs Usual Care for Increasing Screening Uptake in Underscreened Individuals in a US Health Care System, Stratified by Select Patient Characteristics.
Characteristic | Participants, No./total No. (%) | Absolute difference between groups, % (95% CI)b | Intervention vs control, RR (95% CI)c | P value for characteristic-by-group interaction | |
---|---|---|---|---|---|
Intervention (n = 9843a) | Control (n = 9891) | ||||
Age, y | |||||
30-34 | 206/808 (25.5) | 144/794 (18.1) | 7.4 (3.3-11.4) | 1.41 (1.16-1.70) | .08 |
35-39 | 259/932 (27.8) | 191/915 (20.9) | 6.9 (3.0-10.8) | 1.33 (1.13-1.57) | |
40-44 | 338/1194 (28.3) | 240/1185 (20.3) | 8.1 (4.6-11.5) | 1.40 (1.21-1.61) | |
45-49 | 350/1380 (25.4) | 239/1374 (17.4) | 8.0 (4.9-11.0) | 1.46 (1.26-1.69) | |
50-54 | 433/1682 (25.7) | 297/1707 (17.4) | 8.3 (5.6-11.1) | 1.48 (1.30-1.69) | |
55-59 | 478/1938 (24.7) | 300/1943 (15.4) | 9.2 (6.7-11.7) | 1.60 (1.40-1.82) | |
60-64 | 528/1909 (27.7) | 308/1973 (15.6) | 12.0 (9.5-14.6) | 1.77 (1.56-2.01) | |
Raced | |||||
American Indian/Alaska Native | 28/147 (19.0) | 15/145 (10.3) | 8.7 (0.6-16.8) | 1.84 (1.03-3.30) | .51 |
Asian | 247/893 (27.7) | 171/880 (19.4) | 8.2 (4.3-12.1) | 1.42 (1.20-1.69) | |
Black or African American | 119/438 (27.2) | 71/431 (16.5) | 10.7 (5.3-16.1) | 1.65 (1.27-2.14) | |
More than 1 race | 74/285 (26.0) | 49/283 (17.3) | 8.7 (1.9-15.4) | 1.50 (1.08-2.07) | |
Native Hawaiian or other Pacific Islander | 32/151 (21.2) | 22/139 (15.8) | 5.4 (−3.6-14.3) | 1.34 (0.82-2.20) | |
Other | 63/250 (25.2) | 53/235 (22.6) | 2.6 (−5.0-10.3) | 1.12 (0.81-1.54) | |
White | 1975/7018 (28.1) | 1289/7111 (18.1) | 10.0 (8.6-11.4) | 1.55 (1.46-1.65) | |
Unknown | 54/661 (8.2) | 49/667 (7.3) | NA | NA | |
Ethnicityd | |||||
Hispanic | 128/486 (26.3) | 96/480 (20.0) | 6.3 (1.0-11.6) | 1.32 (1.04-1.66) | .22 |
Non-Hispanic | 2403/8710 (27.6) | 1577/8761 (18) | 9.6 (8.4-10.8) | 1.53 (1.45-1.62) | |
Unknown | 61/647 (9.4) | 46/650 (7.1) | NA | NA | |
Length of health plan enrollment, y | |||||
3.4 to <5 | 556/2230 (24.9) | 359/2240 (16.0) | 8.9 (6.6-11.3) | 1.56 (1.38-1.75) | 0.78 |
5 to <10 | 779/3115 (25.0) | 516/3045 (16.9) | 8.1 (6.1-10.1) | 1.48 (1.34-1.63) | |
≥10 | 1257/4498 (27.9) | 844/4606 (18.3) | 9.6 (7.9-11.3) | 1.53 (1.41-1.65) | |
Time since last Papanicolaou test (by length of health plan enrollment) | |||||
Enrolled 3.4 to <5 y | |||||
>3.4 to <5 | 266/704 (37.8) | 202/710 (28.5) | 9.3 (4.5-14.2) | 1.33 (1.14-1.54) | .005 |
No Papanicolaou test | 290/1526 (19.0) | 157/1530 (10.3) | 8.7 (6.3-11.2) | 1.85 (1.55-2.22) | |
Enrolled 5 to <10 y | |||||
>3.4 to <5 | 530/1519 (34.9) | 397/1468 (27.0) | 7.8 (4.6-11.1) | 1.29 (1.16-1.44) | <.001 |
5 to <10 | 119/540 (22.0) | 66/507 (13.0) | 9.0 (4.5-13.6) | 1.69 (1.29-2.23) | |
No Papanicolaou test | 130/1056 (12.3) | 53/1070 (5.0) | 7.4 (5.0-9.7) | 2.49 (1.83-3.38) | |
Enrolled ≥10 y | |||||
>3.4 to <5 | 864/2186 (39.5) | 652/2252 (29.0) | 10.6 (7.8-13.4) | 1.37 (1.26-1.48) | <.001 |
5 to <10 | 272/1143 (23.8) | 151/1182 (12.8) | 11.0 (7.9-14.1) | 1.86 (1.55-2.23) | |
≥10 | 60/475 (12.6) | 23/506 (4.5) | 8.1 (4.6-11.6) | 2.78 (1.75-4.41) | |
No Papanicolaou test | 61/694 (8.8) | 18/666 (2.7) | 6.1 (3.6-8.5) | 3.25 (1.94 − 5.45) | |
US Census block, median household income, $USe | |||||
<25 000 | 38/140 (27.1) | 17/125 (13.6) | 13.5 (4.1-23.0) | 2.00 (1.19-3.34) | .56 |
25 000-49 999 | 506/2107 (24.0) | 326/2115 (15.4) | 8.6 (6.2-11.0) | 1.56 (1.37-1.77) | |
50 000-74 999 | 901/3448 (26.1) | 568/3439 (16.5) | 9.6 (7.7-11.5) | 1.58 (1.44-1.74) | |
75 000-99 999 | 670/2405 (27.9) | 477/2483 (19.2) | 8.7 (6.3-11.0) | 1.45 (1.31-1.61) | |
≥100 000 | 314/1025 (30.6) | 212/1013 (20.9) | 9.7 (5.9-13.5) | 1.46 (1.26-1.70) | |
Unknown | 163/718 (22.7) | 119/716 (16.6) | NA | NA | |
Travel time from home to primary care clinic, minf | |||||
< 10 | 863/3254 (26.5) | 544/3236 (16.8) | 9.7 (7.7-11.7) | 1.58 (1.43-1.74) | .53 |
10 to <20 | 1061/4086 (26.0) | 722/4048 (17.8) | 8.2 (6.3-9.9) | 1.46 (1.34-1.58) | |
20 to <30 | 398/1407 (28.3) | 255/1415 (18.0) | 10.3 (7.2-13.4) | 1.57 (1.37-1.80) | |
≥30 | 249/1004 (24.8) | 186/1072 (17.4) | 7.4 (3.9-11.0) | 1.43 (1.21-1.69) | |
Unknown | 21/92 (22.8) | 12/120 (10.0) | NA | NA | |
Body mass indexg | |||||
<18.5 | 28/109 (25.7) | 19/98 (19.4) | 6.3 (−5.0-17.6) | 1.32 (0.79-2.21) | .63 |
18.5 to 24.9 | 729/2238 (32.6) | 512/2248 (22.8) | 9.8 (7.2-12.4) | 1.43 (1.30-1.58) | |
25 to 29.9 | 676/2168 (31.2) | 467/2220 (21.0) | 10.1 (7.6-12.7) | 1.48 (1.34-1.64) | |
30 to 34.9 | 443/1549 (28.6) | 278/1603 (17.3) | 11.3 (8.3-14.2) | 1.65 (1.44-1.88) | |
35 to 39.9 | 297/1119 (26.5) | 200/1080 (18.5) | 8.0 (4.5-11.5) | 1.43 (1.22-1.68) | |
≥40 | 290/1248 (23.2) | 190/1184 (16) | 7.2 (4.0-10.4) | 1.45 (1.23-1.71) | |
Unknown | 129/1412 (9.1) | 53/1458 (3.6) | NA | NA | |
Tobacco use | |||||
Current | 236/1276 (18.5) | 159/1290 (12.3) | 6.2 (3.4-9.0) | 1.50 (1.25-1.81) | .77 |
Former | 625/2041 (30.6) | 400/2020 (19.8) | 10.8 (8.2-13.5) | 1.55 (1.39-1.73) | |
Never | 1613/5237 (30.8) | 1092/5232 (20.9) | 9.9 (8.3-11.6) | 1.48 (1.38-1.58) | |
Unknown | 118/1289 (9.2) | 68/1349 (5.0) | NA | NA | |
Charlson Comorbidity Indexh | |||||
0 | 2122/7967 (26.6) | 1386/8052 (17.2) | 9.4 (8.2-10.7) | 1.55 (1.46-1.64) | .47 |
1 | 283/1087 (26.0) | 208/1128 (18.4) | 7.6 (4.1-11.1) | 1.41 (1.20-1.66) | |
2 | 106/432 (24.5) | 72/385 (18.7) | 5.8 (0.2-11.5) | 1.31 (1.01-1.71) | |
≥3 | 81/357 (22.7) | 53/326 (16.3) | 6.4 (0.6-12.3) | 1.40 (1.02-1.90) | |
Adherent to guideline-recommended breast cancer screeningi | |||||
No | 396/2551 (15.5) | 198/2587 (7.7) | 7.9 (6.1-9.6) | 2.03 (1.73-2.38) | .002 |
Yes | 868/2221 (39.1) | 586/2301 (25.5) | 13.6 (10.9-16.3) | 1.53 (1.41-1.67) | |
Unknown | 19/158 (12.0) | 14/155 (9.0) | NA | NA | |
Adherent to guideline-recommended colorectal cancer screeningj | |||||
No | 527/3106 (17.0) | 329/3240 (10.2) | 6.8 (5.1-8.5) | 1.67 (1.47-1.90) | .39 |
Yes | 816/1990 (41.0) | 519/1974 (26.3) | 14.7 (11.8-17.6) | 1.56 (1.42-1.71) | |
Unknown | 12/153 (7.8) | 11/135 (8.1) | NA | NA |
Abbreviation: NA, not applicable.
Patient characteristics are not available for 117 (1.2%) participants in the intervention group who opted out of electronic medical record review.
The 95% CIs are from log-binomial regression models with identity link function.
The 95% CIs are from log-binomial regression models with log link function.
Race and ethnicity from electronic medical record data per patient self-report at usual care patient registration via preset multiselect categorical options, with “other” allowing free text entry. The study variable was programmatically categorized into the displayed categories by coding any multiple selections as “more than 1 race.” Manual coding of the "other" category was precluded because institutional review board approval only allowed for individual-level data for the control arm and intervention arm kit returners.
Individual household income data were not available in the EHR; as a proxy, we used median household income calculated at women’s US Census block.
Travel time to primary care clinic was generated with Network Analyst (ArcInfo v 9.1) using geographic centroids of US Census blocks and geocoded street address using women’s home addresses.41
Body mass index is calculated as weight in kilograms divided by height in meters squared.
Generated from an additive index of comorbid conditions.42
Restricted to participants 52 to 64 years old; adherence is based on Healthcare Effectiveness Data and Information Set (HEDIS) definition.43
Restricted to participants 51 to 64 years old; adherence is based on HEDIS definition.44