Abstract
Background
Primary medical care may improve colorectal cancer (CRC) outcomes through increased use of CRC screening tests and earlier diagnosis. We examined the association between primary care utilization and CRC screening, stage at diagnosis, CRC mortality, and all-cause mortality.
Methods
We conducted a retrospective cohort study of patients, aged 67 to 85 years, diagnosed as having CRC between 1994 and 2005 in the Surveillance, Epidemiology, and End Results–Medicare–linked database. Association of the number of visits to primary care physicians (PCPs) in the 3- to 27-month period before the CRC diagnosis and CRC screening, early-stage diagnosis, CRC mortality, and all-cause mortality were examined using multivariable logistic regression and Cox proportional hazards models.
Results
The odds of CRC screening and early-stage diagnosis increased with increasing number of PCP visits (P<.001 for trend). Compared with persons having 0 or 1 PCP visit, patients with 5 to 10 visits had increased odds of ever receiving CRC screening at least 3 months before diagnosis (adjusted odds ratio, 2.60; 95% CI, 2.48-2.72) and early-stage diagnosis (1.35; 1.29-1.42). Persons with 5 to 10 visits had 16% lower CRC mortality (adjusted hazard ratio [AHR], 0.84; 95% CI, 0.80-0.88) and 6% lower all-cause mortality (0.94; 0.91-0.97) compared with persons with 0 or 1 visit.
Conclusions
Medicare beneficiaries with CRC have better outcomes if they have greater utilization of primary care before diagnosis. Revitalization of primary care in the United States may help strengthen the national efforts to reduce the burden of CRC.
Colorectal cancer (CRC) is the fourth most common cause of new cancers and second leading cause of cancer death sin the United States.1 In 2010, there were an estimated 142 570 incident cases and 51 370 deaths from CRC. The 5-yearsurvival rate is greater than 90% for localized CRC, decreasing to 70% for regional disease and 11% for distant stage.1 Screening for CRC prevents cancer, detects early-stage cancer, and decreases cancer-related mortality.2 However,only 59% of American adults aged 50 years or older have ever received a CRC screening test.3 Primary care physicians (PCPs) play a critical role in the delivery of CRC screening by ordering fecal occult blood tests (FOBTs) or referring patients for colon endoscopy. Indeed, a PCP's recommendation is one of the strongest predictors of patients’ adherence to CRC screening.4,5
Population-based studies have found that a higher supply of PCPs is associated with lower incidence of CRC,6,7 earlier CRC stage at diagnosis,8 and lower mortality.8 However, these ecologic studies are limited in that it is not possible to determine whether individuals with better outcomes are the same as those who received care from PCPs. Therefore, the effect of PCPs on CRC outcomes and the degree that it affects stage at diagnosis and mortality are unclear. Understanding the potential effects of PCPs on CRC outcomes is also important because of the anticipated shortage of approximately 44 000 adult PCPs by 2025.9 Current difficulties in accessing primary care by some populations will be further aggravated by the influx of adults needing primary care resulting from expanded national health coverage in the recently enacted health care reform law.10,11
Medicare serves as a universal health insurance system for 39 million elderly Americans.12 However, use of PCP services varies substantially among Medicare beneficiaries, with many seeing multiple specialists at the exclusion of PCPs.13 Not using PCPs may limit opportunities to receive CRC screening and create barriers to achieving national cancer control goals. Therefore, we examined the association between utilization of PCPs and CRC outcomes in Medicare beneficiaries. We hypothesized that Medicare beneficiaries with few or no visits to a PCP would be more likely to have CRC diagnosed at the advanced stage and have higher mortality and that these differences would be explained by differences in previous receipt of CRC screening, defined as having a CRC screening test at least 3 months before cancer diagnosis.
METHODS
DATA SOURCE AND STUDY SAMPLE
This study used a retrospective cohort of persons diagnosed as having CRC within the Surveillance, Epidemiology, and End Results (SEER)–Medicare–linked database14 between 1994 and 2005 (N=225 459). We excluded persons having cancers of the anus, anal canal, or anorectum (n=91); persons diagnosed as having CRC on death certificates or at autopsy (n=2269)15; persons diagnosed as having other cancers before or within 1 year after their primary CRC diagnosis (n=1710); and those eligible for Medicare because of end-stage renal disease (n=808). We also excluded persons older than 85 years because CRC screening is not recommended in this age group (n=14 087).2 To ensure that participants had at least 24 months of Medicare claims before their CRC diagnosis, we excluded those diagnosed as having CRC before age 67 (n=52 166). We further excluded individuals enrolled in Medicare health maintenance organizations with in 2 years before cancer diagnosis through 1 year after diagnosis (n=42 466) because their claims were unavailable. Finally, we excluded patients without continuous Part A and Part B Medicare coverage during this 3-year period (n=28 237). Our final analytic sample included 83 625 persons with CRC. This study was approved by the institutional review boards of the University of South Florida and Beth Israel Deaconess Medical Center.
UTILIZATION OF PRIMARY CARE
To assess primary care services, we examined Medicare claims for the following ambulatory-based evaluation and management services: routine office visits (Current Procedural Terminology16-18 [CPT] codes 99201-99205, 99211-99215, and 99354-99359), outpatient consultations (99241-99245, 99271-99275, 99301-99303, 99311, and 99312), ambulatory visits outside the home (99313, 99315-99316, 993211-99323, 99311-99333, and 99341-99350), and visits for preventive or administrative care (99387, 99397, 99401-99404, 99411-99412, 99420, 99429-99450, and 99455-99546).
We identified the physician specialty associated with each claim by using the unique physician identification number (UPIN) and the Medicare provider specialty field.19,20 We defined PCPs as general practice, family medicine, primary care internal medicine, geriatric medicine, and obstetrics/gynecology (OB/GYN). We included OB/GYN as primary care because 64.3% of visits to OB/GYN practitioners are for routine follow-up or preventive care.21 Because only 2% of claims were from OB/GYN, there was no change in sensitivity analyses that classified an OB/GYN physician as a non-PCP. We identified 160 629 physician UPINs corresponding to primary care specialties and 198 359 physician UPINs corresponding to non–primary care specialties. In addition, there were 7707 physician UPINs (2.1% of all UPINs) with the specialty designated solely as “multi-specialty clinic or group practice.” These were considered non-PCP specialties.
For each participant, we assessed ambulatory care physician claims during the 3- to 27-month period before diagnosis. Because physician visit patterns are likely to change during the time that a potential cancer is being diagnosed, we excluded the 3-month period immediately before diagnosis and assessed physician claims during the 24-month period before this.22 We assessed visits to PCPs by calculating the total number of ambulatory care claims to PCPs during this period. We created categories of PCP visits corresponding to quartiles (0 or 1, 2-4, 5-10, and ≥11 visits). Visits to non-PCPs were assessed in a similar fashion.
STAGE AT DIAGNOSIS AND MORTALITY
Stage at diagnosis was classified using the American Joint Commission on Cancer23,24 staging system (0, I, II, III, and IV), with early-stage CRC defined as stages 0 and I and late stages defined as II, III, and IV. Because distal CRCs are more susceptible to early detection, we created a variable differentiating proximal lesions (proximal to and including the splenic flexure) and distal lesions (the descending colon through the rectum).25
The SEER Program conducts follow-up annually to ascertain vital statistics for all cases. Linkages are made to state vital statistics to obtain the date and underlying cause of death. In all-cause mortality analyses, persons who were alive at the end of follow-up (December 31, 2007) were censored; in CRC mortality, those who died of causes other than CRC were also censored.
CANCER SCREENING TESTS
As one probable mechanism by which primary care would lead to an earlier stage at diagnosis and a lower CRC mortality, we examined whether the number of PCP visits was associated with ever having CRC screening. We assessed the following CPT16 codes; International Classification of Diseases, Ninth Revision (ICD-9) codes; and Healthcare Common Procedure Coding System (HCPCS)26 codes: FOBT (CPT codes 82270 and 82273 and HCPCS code G0107), sigmoidoscopy (CPT codes 45305, 45308, 45309, 45315, 45320, and 45331 and HCPCS code G0104), colonoscopy (CPT codes 45380, 45384, and 45385 HCPCS codes G0105 and G0121), barium enema (CPT codes 74270 and 74280 and HCPCS codes G0106, G0120, and G0122), and office visits for CRC screening (ICD-9 codes V76.51 and V76.41). We included any claim for CRC-related services (FOBT, sigmoidoscopy, colonoscopy, and barium enema) for the full study period, excluding the 3 months before the diagnosis, to better capture receipt of colonoscopies. The mean (SD) time for assessment before CRC screening was 6.7 (3.3) years. Consistent with previous studies of preventive care,22,27,28 we excluded any CRC screening tests in the 3 months before the diagnosis to exclude tests potentially related to the diagnosis of CRC.
STATISTICAL ANALYSIS
The relationship between PCP visits and previous CRC screening was evaluated using multivariable logistic regression. Likewise, we used multivariable logistic regression models to examine the relationship between PCP visits and early-stage diagnosis, excluding persons with unknown stage (n=8049). Odds ratios of early-stage (0 or I) compared with late-stage (II, III, or IV) diagnosis and corresponding 95% CIs were calculated for each category of PCP visits compared with the reference group (0 or 1 visit). To determine whether PCP visits were associated with early-stage diagnosis beyond receipt of previous CRC screening, we fitted logistic models with and without previous CRC screening and examined changes in the estimated odds ratio for PCP visits.
We considered the following potential confounders in multivariable models: number of non-PCP visits, age at diagnosis, sex, race/ethnicity, marital status at diagnosis, census-derived measures of median household income (approximate quin-tiles within each registry), educational levels (approximate quintiles within each registry), metropolitan statistical area, SEER geographic registry (with indicator variables for each registry), Charlson comorbidity index29 (determined from both inpatient and outpatient physician claims), anatomic site (proximal vs distal lesions), and histologic cancer type (adenocarcinomas, including all subtypes, carcinoid tumors, and other). Medicare reimbursement of CRC screening occurred incre mentally, with no coverage from 1994 to 1997, limited coverage between 1998 and 2000 (FOBT yearly, flexible sigmoidos-copy every 4 years, and colonoscopy only for high-risk persons), and full coverage thereafter (colonoscopy every 10 years for all persons). We therefore created indicator variables corresponding to these 3 periods and included them in the models. To account for healthy behaviors in patients seeking more primary care,30 we added receipt of an influenza vacciation in the preceding 3 to 27 months as a proxy for healthy behavior.
We examined CRC-specific mortality and all-cause mortality among persons having invasive CRC (excluding 7732 in situ cancers).8,28 The association between PCP visits and CRC mortality was analyzed using Cox proportional regression models, adjusting for potential confounding factors described in the preceding paragraph in addition to tumor characteristics to control for residual confounding within each stage. To determine whether associations between PCP visits and CRC mortality were primarily the result of previous CRC screening and an earlier stage at diagnosis, models were first performed without previous CRC screening, stage at diagnosis, and tumor characteristics and then repeated including previous CRC screening, stage at diagnosis, and tumor characteristics. Similar analyses were performed to examine the relationship between PCP visits and all-cause mortality. All analyses were performed using commercial software (SAS version 9.2; SAS Institute, Inc, Cary, North Carolina). Data are given as mean (SD) unless otherwise indicated.
RESULTS
Table 1 describes our study cohort. The mean age was 75.8 (6.2) years, and most patients were non-Hispanic white. The mean number of visits to a PCP in the 3- to 27-month period before diagnosis was 7.2 (8.2), with 27.8% of the sample having 0 to 1 visit (25.0% had ≥11 visits). The distribution of visits by specific primary care specialty was general practice, 10.6%; family medicine, 30.5%; internal medicine, 55.9%; OB/GYN, 2.2%; and geriatric medicine, 0.8%. Patients had a mean of 6.9 (9.7) visits to a non-PCP, with 34.1% having 0 to 1 visit (24.7% had ≥11 visits).
Table 1.
Characteristics of 83 625 Patients With Colorectal Cancer
| Characteristic | No. (%) |
|---|---|
| PCP visits in previous 3-27 mo | |
| 0 or 1 | 23 206 (27.8) |
| 2-4 | 20 407 (24.4) |
| 5-10 | 19 140 (22.9) |
| ≥11 | 20 872 (25.0) |
| Non-PCP visits in previous 3-27 mo | |
| 0 or 1 | 28 480 (34.1) |
| 2-4 | 17 480 (20.9) |
| 5-10 | 16 998 (20.3) |
| ≥11 | 20 667 (24.7) |
| Receipt of influenza vaccination in previous 3-27 mo | |
| No | 40 996 (49.0) |
| Yes | 42 629 (51.0) |
| Age at diagnosis, y | |
| 67-75 | 40 330 (48.2) |
| 76-85 | 43 295 (51.8) |
| Sex | |
| Male | 39 821 (47.6) |
| Female | 43 804 (52.4) |
| Race/ethnicity | |
| White, non-Hispanic | 69 191 (82.7) |
| Black, non-Hispanic | 6278 (7.5) |
| Hispanic | 3738 (4.5) |
| Asian/American Indian/Pacific Islander | 4268 (5.1) |
| Other | 150 (0.2) |
| Charlson comorbidity index | |
| 0 | 47 541 (56.9) |
| 1 | 19 766 (23.6) |
| ≥2 | 16 318 (19.5) |
| Year of diagnosis | |
| 1994-1997 | 19 429 (23.2) |
| 1998-2000 | 18 605 (22.3) |
| 2001-2005 | 45 591 (54.5) |
| Marital status | |
| Married, including common law | 6295 (7.5) |
| Single, never married | 44 805 (53.6) |
| Separated/divorced | 4931 (5.9) |
| Widowed | 23 998 (28.7) |
| Unknown | 3596 (4.3) |
| MSA of residence (n = 83 623) | |
| Large metropolitan | 46 423 (55.5) |
| Metropolitan | 22 930 (27.4) |
| Urban | 5407 (6.5) |
| Less urban | 7253 (8.7) |
| Unknown | 1610 (1.9) |
| Educational level (n = 80 281) | |
| Quintile 1, lowest | 15 996 (19.9) |
| Quintile 2 | 15 951 (19.9) |
| Quintile 3 | 16 188 (20.2) |
| Quintile 4 | 15 921 (19.8) |
| Quintile 5, highest | 16 225 (20.2) |
| Income level (n = 83 103) | |
| Quintile 1, lowest | 17 224 (20.7) |
| Quintile 2 | 16 591 (20.0) |
| Quintile 3 | 16 448 (19.8) |
| Quintile 4 | 16 562 (19.9) |
| Quintile 5, highest | 16 278 (19.6) |
| SEER registrya | |
| San Francisco, 1973+ | 3615 (4.3) |
| Connecticut, 1973+ | 8085 (9.7) |
| Detroit, 1973 + | 8443 (10.1) |
| Hawaii, 1973+ | 1558 (1.9) |
| Iowa, 1973+ | 8863 (10.6) |
| New Mexico, 1973+ | 2182 (2.6) |
| Seattle, 1974+ | 4912 (5.9) |
| Utah, 1973+ | 2292 (2.7) |
| Atlanta, 1975+ | 2700 (3.2) |
| San Jose, 1988+ | 2164 (2.6) |
| Los Angeles, 1988+ | 7586 (9.1) |
| Rural Georgia, 1992+ | 268 (0.3) |
| Greater California, 2000+ | 9219 (11.0) |
| Kentucky, 2000+ | 5418 (6.5) |
| Louisiana, 2000+ | 4181 (5.0) |
| New Jersey, 2000+ | 11 176 (13.4) |
| Unknown | 963 (1.2) |
| Histologic type | |
| Adenocarcinoma | 82 177 (98.3) |
| Carcinoid | 974 (1.2) |
| Miscellaneous | 474 (0.6) |
| Stage at diagnosis | |
| 0, In situ | 6913 (8.3) |
| I | 18 624 (22.3) |
| II | 23 028 (27.5) |
| III | 18 389 (22.0) |
| IV | 8622 (10.3) |
| Unknown | 8049 (9.6) |
| Tumor grade | |
| Well differentiated | 7650 (9.2) |
| Moderately differentiated | 49 189 (58.8) |
| Poorly differentiated | 13188 (15.8) |
| Undifferentiated | 667 (0.8) |
| Unknown | 12 931 (15.5) |
| Tumor size, mean (SD), cm | 40.2 (25.0) |
| Tumor location | |
| Rectum | 21 512 (25.7) |
| Colon | 62 113 (74.3) |
| Anatomic site | |
| Distal | 45 276 (54.1) |
| Proximal | 38 349 (45.9) |
Abbreviations: MSA, metropolitan statistical area; PCP, primary care physician; SEER, Surveillance, Epidemiology, and End Results.
The year with the “+” sign refers to the year when the site became part of the SEER registry (from that date to the present).
RECEIPT OF PREVIOUS CRC SCREENING
Overall, 38 220 persons (45.7%) had at least 1 claim of ever receiving a CRC screening test more than 3 months before the CRC diagnosis. Most claims were for FOBT (40.3%), with other services less common (office visit for CRC, 8.6%; colonoscopy, 11.8%; barium enema, 10.3%; and sigmoidoscopy, 2.0%). The likelihood of having at least 1 claim for previous CRC screening increased with increasing number of PCP visits (0 or 1 visit, 27.8%; 2-4 visits, 45.9%; 5-10 visits, 53.4%; and ≥11 visits, 58.3%; P < .001 for trend). Table 2 describes predictors of previous CRC screening. The odds of previous CRC screening increased with increasing number of PCP visits. Previous CRC screening was also independently associated with non-PCP visits, influenza vaccination, older age, female sex, non-Hispanic ethnicity, no comorbidity, later year of diagnosis, being married, and residing in areas with higher educational levels.
Table 2.
Predictors of Prior Colorectal Cancer Screening Test in 83 625 Patientsa
| Characteristic | AORb (95% Wald CI) |
|---|---|
| PCP visits in previous 3-27 mo | |
| 0 or 1 | 1.00 [Reference] |
| 2-4 | 2.07 (1.98-2.16) |
| 5-10 | 2.60 (2.48-2.72) |
| ≥11 | 2.96 (2.83-3.10) |
| Non-PCP visits in previous 3-27 mo | |
| 0 or 1 | 1.00 [Reference] |
| 2-4 | 1.57 (1.51-1.64) |
| 5-10 | 1.87 (1.79-1.95) |
| ≥11 | 2.74 (2.63-2.86) |
| Receipt of influenza vaccination in previous 3-27mo | |
| No | 1.00 [Reference] |
| Yes | 1.53 (1.48-1.58) |
| Age at diagnosis, y | |
| 67-75 | 1.00 [Reference] |
| 76-85 | 1.28 (1.24-1.32) |
| Sex | |
| Male | 1.00 [Reference] |
| Female | 1.26 (1.22-1.30) |
| Race/ethnicity | |
| White, non-Hispanic | 1.00 [Reference] |
| Black, non-Hispanic | 0.96 (0.90-1.02) |
| Hispanic | 0.79 (0.73-0.86) |
| Asian/American Indian/Pacific Islander | 0.80 (0.74-0.86) |
| Other | 0.92 (0.64-1.32) |
| Charlson comorbidity index | |
| 0 | 1.00 [Reference] |
| 1 | 0.95 (0.92-0.99) |
| ≥2 | 0.84 (0.80-0.87) |
| Year of diagnosis | |
| 1994-1997 | 1.00 [Reference] |
| 1998-2000 | 1.41 (1.35-1.47) |
| 2001-2005 | 1.22 (1.17-1.27) |
| Marital status | |
| Married, including common law | 1.00 [Reference] |
| Single, never married | 0.88 (0.83-0.93) |
| Separated/divorced | 0.83 (0.78-0.89) |
| Widowed | 0.87 (0.84-0.90) |
| Unknown | 0.94 (0.87-1.02) |
| MSA of residence | |
| Large metropolitan | 1.00 [Reference] |
| Metropolitan | 0.91 (0.87-0.95) |
| Urban | 0.98 (0.91-1.05) |
| Less urban | 0.84 (0.78-0.91) |
| Unknown | 0.85 (0.75-0.96) |
| Educational level | |
| Quintile 1, lowest | 1.00 [Reference] |
| Quintile 2 | 1.09 (1.04-1.15) |
| Quintile 3 | 1.13 (1.07-1.20) |
| Quintile 4 | 1.22 (1.15-1.29) |
| Quintile 5, highest | 1.39 (1.29-1.49) |
| Income level | |
| Quintile 1, lowest | 1.00 [Reference] |
| Quintile 2 | 0.93 (0.89-0.98) |
| Quintile 3 | 1.01 (0.95-1.06) |
| Quintile 4 | 0.96 (0.90-1.02) |
| Quintile 5, highest | 1.01 (0.94-1.08) |
| SEER registryc | |
| San Francisco, 1973+ | 1.00 [Reference] |
| Connecticut, 1973+ | 1.02 (0.93-1.12) |
| Detroit, 1973+ | 0.72 (0.66-0.78) |
| Hawaii, 1973+ | 0.97 (0.84-1.12) |
| Iowa, 1973+ | 0.82 (0.74-0.91) |
| New Mexico, 1973+ | 0.60 (0.53-0.68) |
| Seattle, 1974+ | 1.02 (0.93-1.13) |
| Utah, 1973+ | 0.59 (0.52-0.67) |
| Atlanta, 1975+ | 0.92 (0.82-1.02) |
| San Jose, 1988+ | 1.02 (0.90-1.15) |
| Los Angeles, 1988+ | 0.79 (0.31-2.02) |
| Rural Georgia, 1992+ | 0.75 (0.69-0.82) |
| Greater California, 2000+ | 0.90 (0.68-1.18) |
| Kentucky, 2000+ | 0.93 (0.85-1.02) |
| Louisiana, 2000+ | 0.72 (0.65-0.80) |
| New Jersey, 2000+ | 0.52 (0.47-0.58) |
| Unknown | 0.63 (0.58-0.68) |
Abbreviations: AOR, adjusted odds ratio; MSA, Metropolitan Statistical Area; PCP, primary care physician; SEER, Surveillance, Epidemiology, and End Results.
Ever received of fecal occult blood test, sigmoidoscopy, barium enema, or colonoscopy more than 3 months before a colorectal cancer diagnosis.
Adjusted for all other variables in the table.
The year with the “+” sign refers to the year when the site became part of the SEER registry (from that date to the present).
STAGE AT CRC DIAGNOSIS
Predictors of early stage at diagnosis (American Joint Commission on Cancer stages 0 and 1) are presented in Table 3. The likelihood of having early-stage CRC diagnosis increased with increasing number of PCP visits (0 or 1 visit, 29.1%; 2-4 visits, 33.2%; 5-10 visits, 36.0%; and ≥11 visits, 37.5%; P<.001 for trend). Compared with persons having 0 or 1 primary care visit, those with 5 to 10 visits had 35% greater odds of receiving an early-stage CRC diagnosis. Early-stage diagnosis was also independently associated with non-PCP visits, influenza vaccination, younger age, male sex, non-Hispanic ethnicity, later year of diagnosis, and distal location of the tumor.
Table 3.
Predictors of Early-Stage Colorectal Cancer Diagnosis in 75 576 Patientsa
| Characteristic | AORb (95%Wald CI) |
|---|---|
| PCP visits in previous 3-27 mo | |
| 0 or 1 | 1.00 [Reference] |
| 2-4 | 1.22 (1.17-1.28) |
| 5-10 | 1.35 (1.29-1.42) |
| ≥11 | 1.41 (1.35-1.48) |
| Non-PCP visits in previous 3-27 mo | |
| 0 or 1 | 1.00 [Reference] |
| 2-4 | 1.19 (1.13-1.24) |
| 5-10 | 1.38 (1.32-1.44) |
| ≥11 | 1.61 (1.53-1.68) |
| Receipt of influenza vaccination in previous 3-27 mo | |
| No | 1.00 [Reference] |
| Yes | 1.12 (1.08-1.15) |
| Age at diagnosis, y | |
| 67-75 | 1.00 [Reference] |
| 76-85 | 0.89 (0.86-0.92) |
| Sex | |
| Male | 1.00 [Reference] |
| Female | 0.96 (0.92-0.99) |
| Race/ethnicity | |
| White, non-Hispanic | 1.00 [Reference] |
| Black, non-Hispanic | 1.03 (0.96-1.10) |
| Hispanic | 0.86 (0.79-0.93) |
| Asian/American Indian/Pacific Islander | 0.94 (0.87-1.03) |
| Other | 1.01 (0.70-1.46) |
| Charlson comorbidity index | |
| 0 | 1.00 [Reference] |
| 1 | 0.98 (0.94-1.02) |
| ≥2 | 1.02 (0.97-1.06) |
| Year of diagnosis | |
| 1994-1997 | 1.00 [Reference] |
| 1998-2000 | 1.15 (1.10-1.21) |
| 2001-2005 | 1.21 (1.16-1.26) |
| Marital status | |
| Married, including common law | 1.00 [Reference] |
| Single, never married | 0.96 (0.90-1.03) |
| Separated/divorced | 0.92 (0.86-0.99) |
| Widowed | 0.91 (0.88-0.95) |
| Unknown | 1.53 (1.41-1.66) |
| MSA of residence | |
| Large metropolitan | 1.00 [Reference] |
| Metropolitan | 1.10 (1.05-1.16) |
| Urban | 1.16 (1.07-1.25) |
| Less urban | 1.09 (1.01-1.18) |
| Unknown | 1.16 (1.02-1.32) |
| Educational level | |
| Quintile 1, lowest | 1.00 [Reference] |
| Quintile 2 | 0.95 (0.90-1.01) |
| Quintile 3 | 0.96 (0.90-1.01) |
| Quintile 4 | 0.94 (0.89-1.01) |
| Quintile 5, highest | 0.98 (0.91-1.06) |
| Income level | |
| Quintile 1, lowest | 1.00 [Reference] |
| Quintile 2 | 1.03 (0.97-1.08) |
| Quintile 3 | 1.06 (1.00-1.12) |
| Quintile 4 | 1.03 (0.96-1.09) |
| Quintile 5, highest | 1.06 (0.98-1.14) |
| SEER registryc | |
| San Francisco, 1973+ | 1.00 [Reference] |
| Connecticut, 1973+ | 1.11 (1.01-1.22) |
| Detroit, 1973+ | 0.95 (0.86-1.04) |
| Hawaii, 1973+ | 1.17 (1.00-1.36) |
| Iowa, 1973+ | 0.94 (0.85-1.05) |
| New Mexico, 1973+ | 0.93 (0.81-1.06) |
| Seattle, 1974+ | 0.84 (0.76-0.94) |
| Utah, 1973+ | 0.98 (0.86-1.12) |
| Atlanta, 1975+ | 1.12 (0.99-1.26) |
| San Jose, 1988+ | 0.95 (0.84-1.08) |
| Los Angeles, 1988+ | 1.14 (1.04-1.25) |
| Rural Georgia, 1992+ | 0.76 (0.56-1.04) |
| Greater California, 2000+ | 0.98 (0.89-1.07) |
| Kentucky, 2000+ | 1.05 (0.94-1.16) |
| Louisiana, 2000+ | 1.01 (0.91-1.13) |
| New Jersey, 2000+ | 1.10 (1.00-1.20) |
| Unknown | 0.75 (0.51-1.12) |
| Anatomic site | |
| Distal | 1.00 [Reference] |
| Proximal | 0.57 (0.55-0.59) |
Abbreviations: AOR, adjusted odds ratio; MSA, Metropolitan Statistical Area; PCP, primary care physician; SEER, Surveillance, Epidemiology, and End Results.
American Joint Commission on Cancer stages 0 and 1.
Adjusted for all other variables in the table.
The year with the “+” sign refers to the year when the site became part of the SEER registry (from that date to the present).
When receipt of previous CRC screening was added to the model, persons having such claims were more likely to be diagnosed as having early-stage cancers (adjusted odds ratio, 1.37; 95% CI, 1.33-1.42). Controlling for previous CRC screening modestly reduced the association of PCP visits and early-stage diagnosis (0 or 1 visit: reference group; 2-4 visits: adjusted odds ratio, 1.16 [95% CI, 1.11-1.22]; 5-10 visits: 1.27 [1.21-1.33]; and ≥11 visits: 1.32 [1.25-1.38]).
CRC AND ALL-CAUSE MORTALITY
Among the 76 712 persons with invasive CRC, there were 43 591 deaths overall and 16 822 deaths from CRC during the follow-up period. Table 4 describes predictors of CRC-specific mortality. In multivariable analyses adjusting for all factors except previous CRC screening, stage at diagnosis, and tumor characteristics, the number of PCP visits was associated with lower CRC mortality. Persons with 5 to 10 PCP visits had 16% lower CRC mortality compared with the reference group. In analyses that further adjusted for CRC screening, stage at diagnosis, and tumor characteristics, 2 to 10 PCP visits remained associated with lower CRC mortality, although the relationship was attenuated.
Table 4.
Predictors of Colorectal Cancer Mortality Among 76 712 Patientsa
| HR (95% CI) |
|||
|---|---|---|---|
| Characteristic | Unadjusted | Multivariable Model 1b | Multivariable Model 2c |
| PCP visits in previous 3-27 mo | |||
| 0 or 1 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 2-4 | 0.85 (0.81-0.88) | 0.89 (0.85-0.92) | 0.96 (0.91-1.00) |
| 5-10 | 0.79 (0.76-0.82) | 0.84 (0.80-0.88) | 0.94 (0.90-0.99) |
| ≥11 | 0.82 (0.78-0.85) | 0.87 (0.83-0.92) | 1.00 (0.95-1.05) |
| Non-PCP visits in previous 3-27 mo | |||
| 0 or 1 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 2-4 | 0.83 (0.80-0.87) | 0.88 (0.84-0.92) | 0.93 (0.89-0.97) |
| 5-10 | 0.73 (0.70-0.76) | 0.78 (0.74-0.81) | 0.87 (0.83-0.91) |
| ≥11 | 0.72 (0.69-0.75) | 0.75 (0.72-0.79) | 0.87 (0.83-0.91) |
| Receipt of influenza vaccination in previous 3-27 mo | |||
| No | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Yes | 0.79 (0.77-0.81) | 0.87 (0.84-0.90) | 0.92 (0.89-0.96) |
| Age at diagnosis, y | |||
| 67-75 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 76-85 | 1.17 (1.14-1.21) | 1.24 (1.20-1.28) | 1.32 (1.28-1.36) |
| Sex | |||
| Male | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Female | 0.98 (0.95-1.01) | 0.93 (0.90-0.96) | 0.95 (0.91-0.98) |
| Race/ethnicity | |||
| White, non-Hispanic | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Black, non-Hispanic | 1.34 (1.27-1.42) | 1.23 (1.16-1.31) | 1.15 (1.08-1.23) |
| Hispanic | 1.03 (0.96-1.11) | 1.00 (0.92-1.08) | 0.95 (0.87-1.03) |
| Asian/American Indian/Pacific Islander | 0.91 (0.84-0.98) | 0.92 (0.84-1.00) | 0.92 (0.84-1.00) |
| Other | 0.92 (0.63-1.33) | 0.84 (0.56-1.26) | 0.73 (0.49-1.09) |
| Charlson comorbidity index | |||
| 0 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 1 | 0.95 (0.91-0.98) | 1.04 (1.00-1.09) | 1.08 (1.04-1.13) |
| ≥2 | 1.02 (0.98-1.06) | 1.18 (1.12-1.23) | 1.25 (1.19-1.31) |
| Year of diagnosis | |||
| 1994-1997 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 1998-2000 | 0.86 (0.83-0.90) | 0.88 (0.84-0.92) | 0.90 (0.86-0.93) |
| 2001-2005 | 0.57 (0.55-0.59) | 0.56 (0.54-0.59) | 0.70 (0.67-0.73) |
| Marital status | |||
| Married, including common law | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Single, never married | 1.25 (1.18-1.32) | 1.14 (1.08-1.22) | 1.02 (1.10-1.24) |
| Separated/divorced | 1.19 (1.12-1.27) | 1.16 (1.08-1.24) | 1.13 (1.05-1.21) |
| Widowed | 1.19 (1.15-1.23) | 1.14 (1.09-1.18) | 1.10 (1.06-1.15) |
| Unknown | 1.03 (0.95-1.11) | 0.91 (0.83-0.99) | 0.88 (0.80-0.96) |
| MSA of residence | |||
| Large metropolitan | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Metropolitan | 1.03 (1.00-1.07) | 0.98 (0.93-1.03) | 1.02 (0.97-1.08) |
| Urban | 1.00 (0.93-1.07) | 0.94 (0.87-1.01) | 0.96 (0.88-1.04) |
| Less urban | 1.02 (0.96-1.08) | 0.94 (0.87-1.01) | 0.93 (0.85-1.01) |
| Unknown | 1.02 (0.91-1.14) | 0.93 (0.82-1.06) | 0.93 (0.81-1.06) |
| Educational level | |||
| Quintile 1, lowest | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Quintile 2 | 1.08 (1.03-1.14) | 0.97 (0.92-1.02) | 0.95 (0.90-1.00) |
| Quintile 3 | 0.98 (0.94-1.03) | 0.98 (0.93-1.04) | 0.95 (0.89-1.01) |
| Quintile 4 | 0.94 (0.89-0.98) | 0.96 (0.90-1.02) | 0.93 (0.87-0.99) |
| Quintile 5, highest | 0.93 (0.89-0.98) | 1.00 (0.93-1.07) | 0.97 (0.90-1.04) |
| Income level | |||
| Quintile 1, lowest | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Quintile 2 | 1.05 (1.00-1.10) | 1.00 (0.95-1.05) | 1.00 (0.95-1.06) |
| Quintile 3 | 0.96 (0.92-1.01) | 0.98 (0.93-1.04) | 1.02 (0.96-1.08) |
| Quintile 4 | 0.94 (0.90-0.99) | 0.99 (0.93-1.05) | 1.02 (0.96-1.09) |
| Quintile 5, highest | 0.89 (0.85-0.93) | 0.93 (0.87-1.00) | 0.98 (0.90-1.05) |
| SEER registryd | |||
| San Francisco, 1973+ | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Connecticut, 973+ | 0.93 (0.85-1.01) | 0.95 (0.86-1.04) | 0.89 (0.81-0.98) |
| Detroit, 1973+ | 1.00 (0.92-1.09) | 0.98 (0.90-1.06) | 0.91 (0.83-0.99) |
| Hawaii, 1973+ | 0.81 (0.71-0.93) | 0.94 (0.80-1.09) | 0.89 (0.76-1.05) |
| Iowa, 1973+ | 0.91 (0.84-0.99) | 0.95 (0.86-1.04) | 0.94 (0.85-1.04) |
| New Mexico, 1973+ | 1.11 (1.00-1.24) | 1.19 (1.05-1.35) | 1.11 (0.98-1.27) |
| Seattle, 1974+ | 0.94 (0.86-1.02) | 0.97 (0.89-1.07) | 0.92 (0.83-1.01) |
| Utah, 1973+ | 1.05 (0.94-1.17) | 1.10 (0.98-1.25) | 1.08 (0.95-1.22) |
| Atlanta, 1975+ | 0.98 (0.88-1.09) | 0.98 (0.88-1.10) | 0.94 (0.84-1.06) |
| San Jose, 1988+ | 0.90 (0.80-1.01) | 0.93 (0.83-1.05) | 0.90 (0.80-1.01) |
| Los Angeles, 1988+ | 0.83 (0.52-1.35) | 1.00 (0.92-1.09) | 0.91 (0.84-1.00) |
| Rural Georgia, 1992+ | 0.97 (0.89-1.06) | 0.93 (0.71-1.22) | 0.98 (0.74-1.29) |
| Greater California, 2000+ | 0.97 (0.75-1.25) | 0.99 (0.90-1.09) | 0.93 (0.84-1.02) |
| Kentucky, 2000+ | 0.73 (0.67-0.79) | 1.19 (1.07-1.32) | 1.20 (1.07-1.34) |
| Louisiana, 2000+ | 0.84 (0.77-0.93) | 1.06 (0.95-1.19) | 1.01 (0.89-1.13) |
| New Jersey, 2000+ | 0.80 (0.72-0.88) | 1.00 (0.91-1.09) | 0.92 (0.84-1.00) |
| Unknown | 0.75 (0.69-0.82) | 1.30 (0.99-1.71) | 0.85 (0.64-1.12) |
| Tumor location | |||
| Colon | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| Rectum | 1.19 (1.15-1.23) | 1.18 (1.14-1.22) | 1.23 (1.19-1.28) |
| Histologic type | |||
| Adenocarcinoma | 1.00 [Reference] | 1.00 [Reference] | |
| Carcinoid | 0.48 (0.40-0.59) | 0.53 (0.42-0.66) | |
| Miscellaneous | 1.35 (1.13-1.62) | 0.93 (0.76-1.13) | |
| Stage at diagnosis | |||
| I | 1.00 [Reference] | 1.00 [Reference] | |
| II | 2.63 (2.44-2.83) | 1.29 (1.18-1.40) | |
| III | 6.80 (6.34-7.29) | 3.54 (3.27-3.84) | |
| IV | 32.66 (30.45-35.02) | 6.47 (5.79-7.23) | |
| Unknown | 5.96 (5.52-6.44) | 1.60 (1.42-1.79) | |
| Tumor grade | |||
| Well differentiated | 1.00 [Reference] | 1.00 [Reference] | |
| Moderately differentiated | 1.58 (1.47-1.69) | 1.09 (1.02-1.18) | |
| Poorly differentiated | 2.76 (2.57-2.96) | 1.50 (1.39-1.62) | |
| Undifferentiated | 3.13 (2.68-3.66) | 1.89 (1.59-2.24) | |
| Unknown | 2.27 (2.10-2.46) | 1.33 (1.22-1.46) | |
| Tumor size, cm | 1.04 (1.04-1.04) | 1.02 (1.02-1.03) | |
| Previous colorectal cancer screening test | |||
| No | 1.00 [Reference] | 1.00 [Reference] | |
| Yes | 0.77 (0.75-0.79) | 0.91 (0.88-0.94) | |
Abbreviations: HR, hazard ratio; MSA, metropolitan statistical area; PCP, primary care physician; SEER, Surveillance, Epidemiology, and End Results.
Excludes in situ tumors.
Model 1 is adjusted for number of PCP visits, number of non-PCP visits, influenza vaccination, age at diagnosis, sex, race/ethnicity, Charlson comorbidity index, year at diagnosis, marital status, residence, educational level, income, SEER Registry, and tumor location.
Model 2 is adjusted for all the variables in model 1 plus histologic type, stage at diagnosis, tumor grade, tumor size, and previous colorectal cancer screening test.
The year with the “+” sign refers to the year when the site became part of the SEER registry (from that date to the present).
In analyses that controlled for all covariates, 2 to 10 PCP visits were associated with reduced all-cause mortality (0 or 1 visit: reference; 2-4 visits: adjusted hazard ratio, 0.94 [95% CI, 0.91-0.97]; 5-10 visits: 0.94 [0.91-0.97]; and ≥11 visits: 1.04 [1.01-1.07]).
EFFECT OF PCP VISITS STRATIFIED BY NON-PCP VISITS
To disentangle the effect of PCP visits from any physician visit, we reanalyzed each outcome stratified by categories of non-PCP visits. Although 10 376 persons (12.4%) had 0 or 1 visit to both a PCP and non-PCP, there was low correlation between visits to PCPs and visits to non-PCPs (Spearman rank correlation coefficient, 0.22; P<.001). For example, 28.2% of patients with ≥11 visits to a non-PCP had 0 or 1 visit to a PCP. Among patients with 5 to 10 visits to a non-PCP, those with 5 to 10 visits to a PCP had 2.18 times increased odds of previous CRC screening, 45% increased odds of early stage at diagnosis, 11% lower CRC mortality, and 7% lower all-cause mortality, compared with patients with 0 or 1 visit to a PCP (Table 5).
Table 5.
Effect of PCP Visits on Outcomes Stratified by Non-PCP Visits in 76 712 Patients
| Non-PCP Visitsa |
||||
|---|---|---|---|---|
| Characteristic | 0 or 1 | 2-4 | 5-10 | ≥11 |
| Adjusted Odds Ratio (95% CI) | ||||
| CRC screeningb | ||||
| PCP visitsa | ||||
| 0 or 1 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 2-4 | 3.55 (3.28-3.85) | 2.08 (1.89-2.30) | 1.84 (1.67-2.03) | 1.26 (1.16-1.37) |
| 5-10 | 4.70 (4.31-5.13) | 2.81 (2.54-3.11) | 2.18 (1.98-2.41) | 1.59 (1.46-1.73) |
| ≥11 | 5.35 (4.86-5.90) | 3.38 (3.03-3.76) | 2.62 (2.37-2.89) | 1.91 (1.77-2.07) |
| CRC stagec | ||||
| PCP visitsa | ||||
| 0 or 1 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 2-4 | 1.37 (1.27-1.47) | 1.31 (1.18-1.46) | 1.31 (1.18-1.46) | 1.31 (1.18-1.46) |
| 5-10 | 1.54 (1.42-1.68) | 1.45 (1.30-1.61) | 1.45 (1.30-1.61) | 1.45 (1.30-1.61) |
| ≥11 | 1.55 (1.41-1.71) | 1.51 (1.35-1.69) | 1.51 (1.35-1.69) | 1.51 (1.35-1.69) |
|
Adjusted Hazard Ratio (95% CI) | ||||
| CRC mortalityd | ||||
| PCP visits a | ||||
| 0 or 1 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 2-4 | 0.96 (0.89-1.02) | 1.03 (0.93-1.14) | 0.87 (0.78-0.98) | 0.97 (0.88-1.08) |
| 5-10 | 0.89 (0.83-0.97) | 1.02 (0.92-1.14) | 0.89 (0.80-1.00) | 1.01 (0.91-1.12) |
| ≥11 | 0.93 (0.85-1.02) | 1.09 (0.97-1.22) | 1.01 (0.90-1.14) | 0.99 (0.90-1.09) |
| All-cause mortalityd | ||||
| PCP visitsa | ||||
| 0 or 1 | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] | 1.00 [Reference] |
| 2-4 | 0.94 (0.90-0.99) | 0.99 (0.92-1.06) | 0.91 (0.85-0.98) | 0.94 (0.88-1.00) |
| 5-10 | 0.89 (0.85-0.94) | 0.99 (0.92-1.06) | 0.93 (0.87-1.00) | 0.98 (0.92-1.04) |
| ≥11 | 1.01 (0.95-1.07) | 1.12 (1.04-1.21) | 1.05 (0.98-1.13) | 1.02 (0.96-1.07) |
Abbreviations: CRC, colorectal cancer; PCP, primary care physician; SEER, Surveillance, Epidemiology, and End Results.
Number of visits in previous 3 to 27 months before the diagnosis.
Adjusted for influenza vaccination, age at diagnosis, sex, race/ethnicity, comorbidity, year at diagnosis, marital status, residence, educational level, income level, and SEER registry.
Adjusted for influenza vaccination, age at diagnosis, sex, race/ethnicity, comorbidity, year at diagnosis, marital status, residence, educational level, income level, SEER registry, and tumor location.
Adjusted for influenza vaccination, age at diagnosis, sex, race/ethnicity, comorbidity, year at diagnosis, marital status, residence, educational level, income level, SEER registry, tumor location, previous CRC screening, stage, histologic type, grade, and tumor size.
COMMENT
Medicare beneficiaries with CRC had better outcomes if they had more primary care utilization before diagnosis. Even within this universally insured population, 28% of Medicare beneficiaries with CRC had no or only 1 con tact with PCPs in the 3 to 27 months before the diagnosis. These patients had increased risk of not ever receiving CRC screening and of having CRC of a more advanced stage, higher CRC mortality, and higher overall mortality. Although earlier CRC stage at diagnosis was associated with receipt of CRC screening,31 the association of earlier stage at diagnosis with higher utilization of primary care was not explained by receipt of CRC screening. In addition, lower CRC mortality with PCP visits was mostly, but not completely, explained by screening and earlier stage at diagnosis. The number of non-PCP visits was also associated with earlier stage and lower mortality, suggesting that access to medical care in general is important for improved CRC outcomes. However, even among patients having many visits to a non-PCP, the effect of a higher number of PCP visits on improved CRC outcomes persisted. This suggests that, regardless of the number of visits to non-PCPs, access to PCPs confers independent and additional benefits. Unfortunately, 28% of Medicare beneficiaries without a PCP report a problem finding such a physician, and 11% report a problem finding a specialist.10
This study confirms others in finding that many Medicare beneficiaries do not use the services of PCPs13 and that low primary care utilization is associated with a lack of CRC screening.32,33 However, controlling for previous CRC screening did not significantly alter the association between PCP visits and early-stage diagnosis. This suggests that factors other than CRC screening mediate the effect of PCP visits on CRC stage. For example, higher use of PCP services may lead to healthier behaviors (eg, eating less red meat, getting more exercise, not smoking, and drinking less alcohol) and use of medicines (eg, aspirin and nonsteroidal anti-inflammatory drugs) that may affect CRC stage.34-36 Alternatively, findings may be explained by unmeasured patient factors, such as healthier habits in patients who seek more frequent primary care. This “healthy user effect” has been described as a potential bias in observational outcome studies whereby healthier individuals are more likely to adhere to medications or use preventive services.30 As a proxy for healthy behavior, we added influenza vaccination in our models but found no appreciable change in results. We also controlled for comorbid illnesses, which were present in more patients with an increased number of visits to PCPs. In addition, we adjusted for socioeconomic status, which is correlated with overall health and health-seeking tendencies.37 However, these adjustments may not fully account for the healthy user effect.
The lower CRC mortality observed among Medicare beneficiaries with higher utilization of primary care appears to be largely the result of earlier-stage diagnosis and receipt of CRC screening. However, controlling for stage and CRC screening did not eliminate the association. This suggests that PCPs may exert influences on CRC mortality beyond CRC screening and earlier-stage diagnosis. For example, PCPs may play a role in promoting healthy behaviors and other preventive services,38 managing comorbid illnesses present in a majority of patients with cancer,39 and coordinating care to prevent medical, medication, and laboratory errors.40 The healthy user effect may also confound the association of PCP visits with mortality. The finding that having 11 or more visits to PCPs slightly increased overall mortality may be the result of more comorbidities in these patients.
It was interesting to find a high frequency of FOBT claims compared with colonoscopy claims. Recent studies41,42 in the general Medicare population have found higher use of colonoscopy than FOBT, whereas an earlier study43 in Medicare patients with CRC confirms our finding that FOBT is most frequently used preceding the peridiagnostic period. One reason for the high frequency of FOBT in our study may be that we defined CRC screening as ever receiving a CRC screening test more than 3 months before diagnosis. However, even when we included only FOBT within the previous 3 to 27 months, the number of patients receiving FOBT was still almost double the number of those receiving colonoscopy. The more likely reason for our high numbers of FOBT is that we included earlier years before reimbursement and increased use of colonoscopy.
Although this research complements previous ecologic studies6-8 examining PCP supply and CRC outcomes, our findings are based on observational data and therefore cannot establish causal relationships. Several potential limitations should be considered when interpret ing our results. First, this study only included persons aged 67 to 85 years having Medicare fee-for-service insurance who were predominantly white, were relatively healthy, and had a relatively high mean number of physician visits. Therefore, findings may not apply to other populations. A subanalysis of persons aged 67 to 75 years (those in whom CRC screening is routinely recommended) yields even stronger associations of the number of PCP visits with CRC outcomes. We were not able to include persons in Medicare health maintenance organizations because they lack claims data. Compared with patients enrolled in Medicare health maintenance organizations, those who had Medicarefee-for-serviceinsurancereportlongerandhigher-quality relationships with their PCPs.44 Second, our study was limited to administrative data contained within the SEER-Medicare database, which omits important patient factors (eg, healthy behaviors, severity of comorbid illness, and medication use) that may be associated with CRC stage at diagnosis or mortality. Third, we did not differentiate CRC tests according to indication (screening vs diagnostic); thus, the rate of previous CRC screening may be lower than the 45.7% found. We excluded any CRC tests performed in the 3 months before diagnosis to exclude those potentially related to the diagnosis of CRC. Fourth, our measure of primary care was limited to number of visits. We did not have detailed information on the content of visits; therefore, it is uncertain what specific aspects of the primary care visit are most important to improve CRC outcomes. We were not able to assess other core primary care attributes, such as first contact, comprehensive care, and coordinated care.45 Finally, it is uncertain whether some Medicare beneficiaries had difficulty accessing PCPs or chose to have no or limited contact with PCPs. Even relatively small ambulatory care copayments decrease the use of outpatient visits and preventive screenings.46,47 Medicare's recent expansion of coverage for preventive care benefits and annual wellness visits should help emphasize the importance of PCP visits and preventive screenings.48 Further research is needed regarding how use of primary care influences CRC stage and mortality and whether our results hold true for other populations and cancers.
This study adds to the mounting evidence of the benefits of primary care in improving health outcomes49 and underscores the importance of adequate access to a PCP, particularly for Medicare beneficiaries. The new health care reform law has provisions to expand primary care training programs and health insurance to all Americans11; however, reorienting the US health care system toward primary care will need more than just increasing the number of primary care trainees or expanding health insurance. Payment reforms to narrow the specialty–primary care payment gap and reward coordination-of-care activities of PCPs are paramount, as are capital investments to improve the primary care infrastructure and paradigm shifts in public perceptions of primary care and patient expectations.50,51 Fortunately, there is a growing movement in the private and public sectors of the United States for the revitalization of primary care.52 This may help strengthen the national efforts on reducing the burden of CRC.
Acknowledgments
Funding/Support: This study was supported by grant RSGHP-08-141-01-CPHPS from the American Cancer Society. The collection of the California cancer incidence data used in this study was supported by the California Department of Public Health as part of the statewide cancer-reporting program mandated by California Health and Safety Code Section 103885; the National Cancer Institute's SEER Program under contract N01-PC-35136 awarded to the Northern California Cancer Center, contract N01-PC-35139 awarded to the University of Southern California, and contract N02-PC-15105 awarded to the Public Health Institute; and the Centers for Disease Control and Prevention's National Program of Cancer Registries, under agreement U55/CCR921930-02 awarded to the Public Health Institute.
Role of the Sponsor: The funding body had no role in the design and conduct of the study; in the collection, management, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.
Footnotes
Author Contributions: Dr Richard Roetzheim has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Ferrante, McCarthy, Gonzalez, Lee, Love-Jackson, and Roetzheim. Acquisition of data: Lee and Roetzheim. Analysis and interpretation of data: Ferrante, McCarthy, Lee, Chen, and Roetzheim. Drafting of the manuscript: Ferrante and Roetzheim. Critical revision of the manuscript for important intellectual content: Ferrante, Gonzalez, Lee, Chen, and Roetzheim. Statistical analysis: McCarthy, Lee, and Chen. Obtained funding: Roetzheim. Administrative, technical, and material support: Ferrante, McCarthy, Chen, Love-Jackson, and Roetzheim. Study supervision: Roetzheim.
Financial Disclosure: None reported.
Additional Contributions: The authors acknowledge the efforts of the Applied Research Program, NCI; the Office of Research, Development and Information, CMS; Information Management Services, Inc; and the SEER Program tumor registries in the creation of the SEER-Medicare database.
Previous Presentations: A portion of this study was presented as a poster at Academy Health's 2011 Annual Research Meeting; June 13, 2011; Seattle, Washington, and will also be presented at the 39th Annual Meeting of the North American Primary Care Research Group; November 15, 2011; Banff, Alberta, Canada.
Disclaimer: The interpretation and reporting of the SEER-Medicare data are the sole responsibility of the authors. The ideas and opinions expressed herein are those of the authors; endorsement by the State of California Department of Public Health, the National Cancer Institute, and the Centers for Disease Control and Prevention or their contractors and subcontractors is not intended and should not be inferred.
REFERENCES
- 1.Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA Cancer J Clin. 2010;60(5):277–300. doi: 10.3322/caac.20073. [DOI] [PubMed] [Google Scholar]
- 2.US Preventive Services Task Force Screening for colorectal cancer: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2008;149(9):627–637. doi: 10.7326/0003-4819-149-9-200811040-00243. [DOI] [PubMed] [Google Scholar]
- 3.Trends Progress Report C. 2009/2010 Update. National Cancer Institute, NIH, DHHS; Bethesda, MD: Apr, 2010. [June 20, 2011]. 2010. http://progressreport.cancer.gov. [Google Scholar]
- 4.Klabunde CN, Schenck AP, Davis WW. Barriers to colorectal cancer screening among Medicare consumers. Am J Prev Med. 2006;30(4):313–319. doi: 10.1016/j.amepre.2005.11.006. [DOI] [PubMed] [Google Scholar]
- 5.Wee CC, McCarthy EP, Phillips RS. Factors associated with colon cancer screening: the role of patient factors and physician counseling. Prev Med. 2005;41(1):23–29. doi: 10.1016/j.ypmed.2004.11.004. [DOI] [PubMed] [Google Scholar]
- 6.Hao Y, Jemal A, Zhang X, Ward EM. Trends in colorectal cancer incidence rates by age, race/ethnicity, and indices of access to medical care, 1995-2004 (United States). Cancer Causes Control. 2009;20(10):1855–1863. doi: 10.1007/s10552-009-9379-y. [DOI] [PubMed] [Google Scholar]
- 7.Roetzheim RG, Gonzalez EC, Ramirez A, Campbell R, van Durme DJ. Primary care physician supply and colorectal cancer. J Fam Pract. 2001;50(12):1027–1031. [PubMed] [Google Scholar]
- 8.Roetzheim RG, Pal N, Gonzalez EC, et al. The effects of physician supply on the early detection of colorectal cancer. J Fam Pract. 1999;48(11):850–858. [PubMed] [Google Scholar]
- 9.Colwill JM, Cultice JM, Kruse RL. Will generalist physician supply meet demands of an increasing and aging population? Health Aff (Millwood) 2008;27(3):w232–w241. doi: 10.1377/hlthaff.27.3.w232. [DOI] [PubMed] [Google Scholar]
- 10.Medicare Payment Advisory Commission . Report to the Congress: Medicare Payment Policy, March 2009. MedPac; Washington, DC: 2009. [Google Scholar]
- 11.Project HOPE—The People-to-People Health Foundation Inc The end of the beginning: enactment of health reform. Health Aff (Millwood) 2010;29(5):758–759. doi:10.1377/hlthaff.2010.0411. [Google Scholar]
- 12.Medicare Chartbook; 2010. 4th ed. Kaiser Family Foundation; Menlo Park, CA: 2010. [Google Scholar]
- 13.Starfield B, Chang HY, Lemke KW, Weiner JP. Ambulatory specialist use by non-hospitalized patients in US health plans: correlates and consequences. J Ambul Care Manage. 2009;32(3):216–225. doi: 10.1097/JAC.0b013e3181ac9ca2. [DOI] [PubMed] [Google Scholar]
- 14.Warren J, Klabunde C, Schrag D, Bach P, Riley G. Overview of the SEER-Medicare data: content, research applications, and generalizability to the United States elderly population. Med Care. 2002;40((8)(suppl)):VI–3-VI-18. doi: 10.1097/01.MLR.0000020942.47004.03. [DOI] [PubMed] [Google Scholar]
- 15.Boer R, Ries L, van Ballegooijen M, Feuer EJ, Legler J, Habbema D. Ambiguities in Calculating Cancer Patient Survival: the SEER Experience for Colorectal and Prostate Cancer. National Cancer Institute; Washington, DC: 2002. Statistical Research and Applications Branch, NCI, Technical Report No. 2002-05. [Google Scholar]
- 16.Kirschner CG, Burkett RC, Coy JA, et al. American Medical Association. Physicians’ Current Procedural Terminology: CPT '94. American Medical Association; Chicago, IL: 1994. [Google Scholar]
- 17.Kirschner CG, Davis SJ, Evans D, et al. American Medical Association . Current Procedural Terminology: CPT 1999. American Medical Association; Chicago, IL: 1999. [Google Scholar]
- 18.Beebe M, Green G, Pavloski D, et al. American Medical Association . Current Procedural Terminology: CPT 2005. American Medical Association; Chicago, IL: 2004. [Google Scholar]
- 19.Baldwin LM, Adamache W, Klabunde CN, Kenward K, Dahlman C, Warren JL. Linking physician characteristics and Medicare claims data: issues in data availability, quality, and measurement. Med Care. 2002;((8)(suppl)):40, IV–82-IV-95. doi: 10.1097/00005650-200208001-00012. [DOI] [PubMed] [Google Scholar]
- 20.Yu X, McBean AM, Virnig BA. Physician visits, patient comorbidities, and mammography use among elderly colorectal cancer survivors. J Cancer Surviv. 2007;1(4):275–282. doi: 10.1007/s11764-007-0037-7. [DOI] [PubMed] [Google Scholar]
- 21.Valderas JM, Starfield B, Forrest CB, Sibbald B, Roland M. Ambulatory care provided by office-based specialists in the United States. Ann Fam Med. 2009;7(2):104–111. doi: 10.1370/afm.949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Gornick ME, Eggers PW, Riley GF. Associations of race, education, and patterns of preventive service use with stage of cancer at time of diagnosis. Health Serv Res. 2004;39(5):1403–1427. doi: 10.1111/j.1475-6773.2004.00296.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gross CP, Andersen MS, Krumholz HM, McAvay GJ, Proctor D, Tinetti ME. Relation between Medicare screening reimbursement and stage at diagnosis for older patients with colon cancer. JAMA. 2006;296(23):2815–2822. doi: 10.1001/jama.296.23.2815. [DOI] [PubMed] [Google Scholar]
- 24.Lang K, Korn JR, Lee DW, Lines LM, Earle CC, Menzin J. Factors associated with improved survival among older colorectal cancer patients in the US: a population-based analysis. BMC Cancer. 2009;9:227. doi: 10.1186/1471-2407-9-227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gonzalez EC, Ferrante JM, Van Durme DJ, Pal N, Roetzheim RG. Comorbid illness and the early detection of cancer. South Med J. 2001;94(9):913–920. [PubMed] [Google Scholar]
- 26.HCPCS general information. Centers for Medicare & Medicaid Services Web site; [August 16, 2011]. https://www.cms.gov/medhcpcsgeninfo/. [Google Scholar]
- 27.McCarthy EP, Burns RB, Coughlin SS, et al. Mammography use helps to explain differences in breast cancer stage at diagnosis between older black and white women. Ann Intern Med. 1998;128(9):729–736. doi: 10.7326/0003-4819-128-9-199805010-00005. [DOI] [PubMed] [Google Scholar]
- 28.Schonberg MA, Marcantonio ER, Ngo L, Li D, Silliman RA, McCarthy EP. Causes of death and relative survival of older women after a breast cancer diagnosis. J Clin Oncol. 2011;29(12):1570–1577. doi: 10.1200/JCO.2010.33.0472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383. doi: 10.1016/0021-9681(87)90171-8. [DOI] [PubMed] [Google Scholar]
- 30.Brookhart MA, Patrick AR, Dormuth C, et al. Adherence to lipid-lowering therapy and the use of preventive health services: an investigation of the healthy user effect. Am J Epidemiol. 2007;166(3):348–354. doi: 10.1093/aje/kwm070. [DOI] [PubMed] [Google Scholar]
- 31.Fazio L, Cotterchio M, Manno M, McLaughlin J, Gallinger S. Association between colonic screening, subject characteristics, and stage of colorectal cancer. Am J Gastroenterol. 2005;100(11):2531–2539. doi: 10.1111/j.1572-0241.2005.00319.x. [DOI] [PubMed] [Google Scholar]
- 32.Fenton JJ, Reid RJ, Baldwin LM, Elmore JG, Buist DS, Franks P. Influence of primary care use on population delivery of colorectal cancer screening. Cancer Epidemiol Biomarkers Prev. 2009;18(2):640–645. doi: 10.1158/1055-9965.EPI-08-0765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Cardarelli R, Thomas JE. Having a personal health care provider and receipt of colorectal cancer testing. Ann Fam Med. 2009;7(1):5–10. doi: 10.1370/afm.904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Wei EK, Giovannucci E, Wu K, et al. Comparison of risk factors for colon and rectal cancer. Int J Cancer. 2004;108(3):433–442. doi: 10.1002/ijc.11540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Benedetti AL, Collet JP, Boivin JF, Hanley JA. Effect of nonsteroidal anti-inflammatory drugs on stage of colon cancer at diagnosis. J Clin Epidemiol. 2003;56(8):782–787. doi: 10.1016/s0895-4356(03)00131-8. [DOI] [PubMed] [Google Scholar]
- 36.Flossmann E, Rothwell PM. British Doctors Aspirin Trial and the UK-TIA Aspirin Trial. Effect of a spirin on long-term risk of colorectal cancer: consistent evidence from randomised and observational studies. Lancet. 2007;369(9573):1603–1613. doi: 10.1016/S0140-6736(07)60747-8. [DOI] [PubMed] [Google Scholar]
- 37.Humphrey LL, Chan BK, Sox HC. Postmenopausal hormone replacement therapy and the primary prevention of cardiovascular disease. Ann Intern Med. 2002;137(4):273–284. doi: 10.7326/0003-4819-137-4-200208200-00012. [DOI] [PubMed] [Google Scholar]
- 38.Ferrante JM, Balasubramanian BA, Hudson SV, Crabtree BF. Principles of the patient-centered medical home and preventive services delivery. Ann Fam Med. 2010;8(2):108–116. doi: 10.1370/afm.1080. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Ogle KS, Swanson GM, Woods N, Azzouz F. Cancer and comorbidity: redefining chronic diseases. Cancer. 2000;88(3):653–663. doi: 10.1002/(sici)1097-0142(20000201)88:3<653::aid-cncr24>3.0.co;2-1. [DOI] [PubMed] [Google Scholar]
- 40.Starfield B. Reinventing primary care: lessons from Canada for the United States. Health Aff (Millwood) 2010;29(5):1030–1036. doi: 10.1377/hlthaff.2010.0002. [DOI] [PubMed] [Google Scholar]
- 41.Cooper GS, Doug Kou T. Underuse of colorectal cancer screening in a cohort of Medicare beneficiaries. Cancer. 2008;112(2):293–299. doi: 10.1002/cncr.23176. [DOI] [PubMed] [Google Scholar]
- 42.Doubeni CA, Laiyemo AO, Young AC, et al. Primary care, economic barriers to health care, and use of colorectal cancer screening tests among Medicare enrollees over time. Ann Fam Med. 2010;8(4):299–307. doi: 10.1370/afm.1112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Cooper GS, Payes JD. Receipt of colorectal testing prior to colorectal carcinoma diagnosis. Cancer. 2005;103(4):696–701. doi: 10.1002/cncr.20839. [DOI] [PubMed] [Google Scholar]
- 44.Safran DG, Wilson IB, Rogers WH, Montgomery JE, Chang H. Primary care quality in the Medicare Program: comparing the performance of Medicare health maintenance organizations and traditional fee-for-service medicare. Arch Intern Med. 2002;162(7):757–765. doi: 10.1001/archinte.162.7.757. [DOI] [PubMed] [Google Scholar]
- 45.Starfield B. Primary Care: Concept, Evaluation, and Policy. Oxford University Press; New York, NY: 1992. [Google Scholar]
- 46.Trivedi AN, Moloo H, Mor V. Increased ambulatory care copayments and hospitalizations among the elderly. N Engl J Med. 2010;362(4):320–328. doi: 10.1056/NEJMsa0904533. [DOI] [PubMed] [Google Scholar]
- 47.Trivedi AN, Rakowski W, Ayanian JZ. Effect of cost sharing on screening mammography in Medicare health plans. N Engl J Med. 2008;358(4):375–383. doi: 10.1056/NEJMsa070929. [DOI] [PubMed] [Google Scholar]
- 48.Hughes C. What you need to know about the Medicare preventive services expansion. Fam Pract Manag. 2011;18(1):22–25. [PubMed] [Google Scholar]
- 49.Friedberg MW, Hussey PS, Schneider EC. Primary care: a critical review of the evidence on quality and costs of health care. Health Aff (Millwood) 2010;29(5):766–772. doi: 10.1377/hlthaff.2010.0025. [DOI] [PubMed] [Google Scholar]
- 50.Bodenheimer T, Grumbach K, Berenson RA. A lifeline for primary care. N Engl J Med. 2009;360(26):2693–2696. doi: 10.1056/NEJMp0902909. [DOI] [PubMed] [Google Scholar]
- 51.Phillips RL, Jr, Bazemore AW. Primary care and why it matters for U.S. health system reform. Health Aff (Millwood) 2010;29(5):806–810. doi: 10.1377/hlthaff.2010.0020. [DOI] [PubMed] [Google Scholar]
- 52.Grundy P, Hagan KR, Hansen JC, Grumbach K. The multi-stakeholder movement for primary care renewal and reform. Health Aff (Millwood) 2010;29(5):791–798. doi: 10.1377/hlthaff.2010.0084. [DOI] [PubMed] [Google Scholar]