Abstract
Introduction
Patients with multiple chronic conditions (multimorbidity) are commonly seen in primary care practices and often have suboptimal uptake of preventive care owing to competing treatment demands. The complexity of multimorbidity patterns and their impact on receiving preventive services is not fully understood. This study identifies multimorbidity combinations associated with low receipt of preventive services.
Methods
This was a retrospective cohort study of U.S. community health center patients aged ≥19 years. Electronic health record data from 209 community health centers for the January 1, 2014–December 31, 2017 study period were analyzed in 2018–2019. Multimorbidity patterns included: “physical only,” “mental health only,” and “physical and mental health” multimorbidity patterns, with “no multimorbidity” as a reference category. Electronic health record–based preventive ratios (number of months services were up-to-date/total months the patient was eligible for services) were calculated for the 14 preventive services. Negative binomial regression models assessed the relationship between multimorbidity physical/mental health patterns and the preventive ratio for each service.
Results
There was a variation in receipt of preventive care between multimorbidity groups: Individuals with “mental health only” multimorbidity were less likely to be up-to-date with cardiometabolic and cancer screenings compared with the “no multimorbidity” group or groups with physical health conditions, and the “physical only” multimorbidity group had low rates of depression screening.
Conclusions
This study provided critical insights into receipt of preventive service among adults with multimorbidity using a more precise method for measuring up-to-date preventive care delivery. Findings would be useful to identify target populations for future intervention programs to improve preventive care.
INTRODUCTION
Preventive care is critical in improving population health, and can potentially decrease long-term healthcare spending.1 However, only 8% of U.S. adults aged 35 years and older receive all recommended preventive services.2 There is a growing number of people living with multiple chronic conditions—defined as having two or more concurrent chronic conditions, and referred to as multimorbidity.3 Competing treatment demands may lead to suboptimal uptake of preventive care among individuals with multimorbidity.4–6 However, previous studies show inconsistent findings regarding the impact of chronic conditions on utilization of preventive care: Some showed a negative association with receipt of preventive care,7,8 whereas others found a positive relationship even after controlling for age and number of visits.9–11 Different methods to define multimorbidity and type and number of conditions included contribute to such inconsistency.
Mental health conditions pose additional challenges to adequate preventive care provision,12,13 especially with the fragmentation of care in the U.S. healthcare system.14 Individuals with mental health conditions are likely to have more chronic conditions,15,16 and the evidence concerning appropriate receipt of preventive services among patients with mental illness is mixed.17 Some studies have shown that people with mental health conditions are less likely to receive preventive care,12,18,19 whereas others have found no differences in receipt of preventive services between patients with or without mental health conditions.20
Additionally, it is unclear how specific combinations or patterns of chronic conditions may impede the provision of high-quality, appropriate preventive services. More evidence is needed to understand utilization of preventive care by multimorbidity patterns, as multimorbidity poses unique challenges to patients21,22 and care providers,23 and may affect adequate receipt of preventive services.
Moreover, to better understand the impact of competing demands from different combination of physical and mental health conditions, more nuanced measures of preventive care receipt are needed. For instance, preventive ratios estimate the proportion of time a patient was “up-to-date” for a given preventive service.24 This measure was used previously to assess delivery of preventive care in community health centers (CHCs).24,25
Therefore, the purpose of this study is to evaluate receipt of preventive services overall and by types of services among patients with different combinations of physical and mental health conditions compared with no multimorbidity group. Authors hypothesize that individuals with “mental health only” multimorbidity will be less up-to-date with preventive services compared with those with no multimorbidity, and presence of physical health comorbidity will be associated with higher preventive care.
METHODS
The authors conducted a retrospective cohort study of adults, aged ≥19 years, receiving care at one of 209 clinics from 59 CHCs in the U.S. between January 1, 2014 and December 31, 2017. Low-income and uninsured individuals are at higher risk of having multimorbidity26 and comprise a large proportion of CHC patients.27 The authors utilized electronic health record (EHR) data from the OCHIN, Inc. community health information network. OCHIN is a nonprofit organization that provides a fully hosted instance of the Epic® EHR to safety net clinics.28–30 Detailed information about OCHIN and its EHR database is available elsewhere.28,29 Clinics that were not “live” on the EHR throughout the study period or that were unlikely to provide preventive care services to adult patients, and those providing <100 total ambulatory visits or serving a population with <20% adults, were excluded. Data were analyzed in 2018–2019.
Study Population
Among eligible CHCs, authors included patients with two or more ambulatory visits: One or more visits prior to January 1, 2014 were needed to identify “established” patients with a documented medical history to assess chronic condition at the beginning of the study, and one or more visits during the 2014–2017 period.
Measures
The authors used preventive ratios to estimate the proportion of time a patient was up-to-date for a given preventive service.24,25 A preventive ratio is the total person-months covered in a given time period divided by the total person-months eligible for a particular service (Figure 1). For example, if an influenza vaccination was administered to a patient on the day it was due, that individual was considered up-to-date for this service 100% of the eligible time (1 year). The preventive ratio ranges from 0% to 100%. In situations of competing demand, delivery of preventive care might be delayed, thus the preventive ratio can more accurately capture the timeliness of service receipt. For example, if a patient received an influenza vaccination 3 months after it was due, a binary measure would report that the patient received the vaccination while the preventive ratio will indicate that the person was “up-to-date” 9 of the 12 months (75%).
Figure 1.
Definition of the preventive ratio.
Source: Hatch BA et al. Use of a Preventive Index to Examine Clinic-Level Factors Associated With Delivery of Preventive Care. Am J Prev Med. 2019;57(2):241–249.
The authors measured preventive ratios for 14 services based on U.S. Preventive Services Task Force grade A and B recommendations and Centers for Disease Control and Prevention immunization recommendations.31,32 They also used a combined preventive index, which aggregates the average of all individual preventive ratios to assess overall receipt of preventive services.24,25 Services due and received were identified through diagnosis and procedure codes, lab/imaging/scanned test results, and longitudinal “health maintenance” records. Services that were ordered but not verified as received were not counted. For each preventive service, individuals were identified as “due” based on sex and age eligibility. Detailed information about measure-by-measure definitions is available elsewhere.25
Chronic conditions are from historical problem list records and medical diagnosis codes prior to January 1, 2014. The list of chronic conditions was based on Centers for Medicare and Medicaid Services Chronic Conditions Data Warehouse algorithm33 and included 29 chronic conditions categories with an individual prevalence of ≥1% in the study population to increase the epidemiological interest.34 Mental health conditions included: depression, anxiety/post-traumatic stress disorder, attention-deficit/hyperactivity disorder, alcohol use disorder, substance use disorder, group with mental disabilities, and other mental health conditions group. A full list of physical chronic conditions can be found in Table 1 and Appendix Table 1. Individuals were categorized into four mutually exclusive groups: “no multimorbidity” group (zero to one chronic condition), “physical only” multimorbidity group (two or more physical chronic conditions), “mental health only” multimorbidity group (two or more mental health diseases, and no physical comorbidity), and “physical and mental health” multimorbidity group.
Table 1.
Characteristics of the Study Patients, Overall and by Group
| Characteristic | All patients | 0–1 chronic conditions | ≥2 chronic conditions |
||
|---|---|---|---|---|---|
| Physical health only | Mental health only | Physical and mental health | |||
| Study population, n (%) | 301,665 | 142,872 (47.4) | 54,556 (18.1) | 12,282 (4.1) | 91,955 (30.5) |
| Age, years, mean (SD) | 45 (16) | 38 (14) | 57 (15) | 36 (12) | 49 (14) |
| Female, n (%) | 188,500 (62.5) | 92,281 (64.6) | 31,904 (58.5) | 7,594 (61.5) | 56,721 (61.7) |
| Race/Ethnicity, n (%) | |||||
| Non-Hispanic white | 165,476 (54.9) | 68,988 (48.3) | 27,604 (50.6) | 8,787 (71.1) | 60,097 (65.4) |
| Non-Hispanic black | 41,439 (13.7) | 18,499 (13.0) | 9,684 (17.8) | 1,093 (8.9) | 12,163 (13.2) |
| Hispanic | 72,951 (24.2) | 44,102 (30.9) | 13,230 (24.3) | 1,736 (14.1) | 13,883 (15.1) |
| Other | 21,799 (7.2) | 11,283 (7.9) | 4,038 (7.4) | 741 (6.0) | 5,737 (6.2) |
| English-language preference, n (%) | 23,3441 (77.4) | 102,455 (71.7) | 4,0697 (74.6) | 11,100 (89.8) | 79,189 (86.2) |
| Insurance status, n (%) | |||||
| Medicaid | 90,304 (29.9) | 42,149 (29.5) | 10,607 (19.4) | 5,400 (43.7) | 32,148 (35.0) |
| Medicare | 4,7241 (15.7) | 7,730 (5.4) | 16,779 (30.8) | 927 (7.5) | 21,805 (23.7) |
| Private | 51,559 (17.1) | 29,139 (20.4) | 9,817 (18.0) | 1,340 (10.8) | 11,263 (12.3) |
| Uninsured | 99,815 (33.1) | 56,292 (39.4) | 15,303 (28.1) | 4,202 (34.0) | 24,018 (26.1) |
| Other | 12,746 (4.2) | 7,562 (5.3) | 2,050 (3.8) | 488 (4.0) | 2,646 (2.9) |
| Rural location, n (%) | 86,004 (28.5) | 42,076 (29.5) | 17,153 (31.4) | 2,889 (23.4) | 23,886 (26.0) |
| Number of chronic conditions, mean (SD) | 2.2 (2.2) | 0.4 (0.5) | 3.2 (1.4) | 2.5 (0.9) | 4.4 (2.0) |
| Smoking status, n (%) | |||||
| Never | 151,627 (50.3) | 84,246 (59.0) | 33,175 (60.8) | 3,456 (28.0) | 30,750 (33.5) |
| Current | 75,290 (25.0) | 25,522 (17.9) | 5,233 (9.6) | 6,399 (51.8) | 38,136 (41.5) |
| Former | 58,628 (19.4) | 19,747 (13.8) | 15,057 (27.6) | 2,098 (17.0) | 21,726 (23.7) |
| Unknown | 16,120 (5.3) | 13,357 (9.4) | 1,091 (2.0) | 404 (3.3) | 1,268 (1.4) |
| Number of office visits during 2014‒ 2017, mean (SD) | 12.9 (22.5) | 9.2 (12.9) | 14.6 (16.1) | 13.3 (26.7) | 17.7 (33.3) |
| Having PCP by January 1, 2014 | 259,442 (86.0) | 109,892 (76.9) | 50,250 (92.1) | 11,433 (92.5) | 87,867 (95.6) |
| Usual provider continuity index over 2014‒2017, mean (SD) | 50.5 (34.4) | 41.2 (35.9) | 61.5 (30.2) | 47.4 (33.5) | 58.9 (30.0) |
Note: Chronic conditions were defined using the Chronic Condition Data Warehouse (CCW) algorithm created by the Center for Medicare and Medicaid Services. Physical conditions include: hypertension, hyperlipidemia, obesity, diabetes, ulcer/gastroesophageal reflux disease, chronic pain, asthma, arthritis, thyroid disease, anemia, cancer, migraine headache, chronic obstructive pulmonary disease, chronic heart disease, chronic kidney disease, hepatitis, liver disease cerebrovascular disease, vision impairment, hearing Impairment, dysrhythmia, HIV/AIDS. Mental conditions include: depression, anxiety/PTSD, substance abuse, alcohol-related disease/abuse, attention deficit hyperactivity disorder, other mental health diagnosis group, and mental disability group.
“Unknown” category for Race/Ethnicity variable were included in “Other” category given low number of observations.
PCP, primary care provider; PTSD, post-traumatic stress disorder.
Covariates shown to affect preventive care delivery included35–37: patients’ sex, race/ethnicity, preferred spoken language, age as of the study start, insurance type, rural location, smoking status, BMI, established with a primary care provider, number of ambulatory visits during the study period, and continuity of care. Missing/unknown data were included an “unknown” category in the analysis. Continuity of care was measured with the usual provider continuity index,38 which is the percentage of visits with the established primary care provider among all office visits during study period. The authors did not adjust for number of chronic conditions as it would have been correlated with the main independent variable (e.g., no multimorbidity group would be perfectly correlated with zero chronic conditions) and would have produced unstable regression coefficients. The authors did not differentiate groups by number of conditions but instead evaluated different multimorbidity patterns.
Statistical Analysis
Frequencies and percentages for categorical variables and mean and SD for continuous variables for each group were calculated. The authors calculated unadjusted mean and SD of the preventive ratio for individual preventive services and combined preventive index among the four groups. The mean preventive ratio represents the average percentage of months study patients were “covered” for a given service, of the total months eligible, from 2014 to 2017. To assess the association between different multimorbidity patterns and preventive ratios, they utilized negative binomial regression models accounting for overdispersion.
The authors then performed univariable and multivariable models for each preventive service, controlling for EHR-based covariates. They did not control for clinic-level characteristics because they believe that the rich collection of patient-level characteristics included in the model already accounted for heterogeneity across clinics. Recognizing that preventive care delivery varies by practice setting,39 they accounted for nesting of patients within clinics by robust cluster-adjusted SE adjustment. The prevalence of multimorbidity combinations varies by age group,40 thus for sensitivity analysis the authors performed regression models for each age group individually. In addition, they performed regression models with different reference categories such as “physical only” multimorbidity group and “no chronic condition” group to compare the “mental health only” group to groups without any mental health diagnoses. Unadjusted and adjusted rate ratios and 95% CIs were calculated. Statistical significance was considered at the α<0.05 level. All analyses were conducted using Stata, version 15. This study was approved by the Oregon Health & Science University IRB.
RESULTS
Overall, the study sample included 301,665 patients, with mean age of 45 years, 62.5% women, 29.9% with Medicaid insurance and 33.1% uninsured (Table 1). More than 50% of patients had two or more chronic conditions. About 40% had one or more mental health conditions, and among patients with multimorbidity, mental health diagnoses were prevalent (65%). A “physical only” pattern was identified in 18.1% of patients; 30.5% had a combination of physical and mental health multimorbidity, and 4.1% had “mental health only” pattern. Table 1 shows the characteristics of the four study groups. Patients in “physical only” and “physical and mental health” groups tended to be older and had more chronic conditions and office visits than the “no multimorbidity” and the “mental health only” groups. Individuals in the “mental health only” group were more likely to be white, have Medicaid insurance coverage, and be a current smoker than individuals in the other groups.
Table 2 shows percentage of person-time covered for each of the services across different multimorbidity patterns. Overall, the combined preventive index was <50% and varied from 33.9% (“no multimorbidity” group) to 46.3% (“physical only” group). Patients with no multimorbidity had lower preventive ratios compared with study groups with physical health conditions, except in chlamydia screening (31.0%). Patients with “mental health only” pattern had lower preventive ratios for diabetes and lipid screenings, all cancer screenings, chlamydia and abdominal aortic aneurysm screening, and pneumonia vaccination compared with the “no multimorbidity” group.
Table 2.
Eligible Patients for Each Preventive Service and Percentage of Patient-Time Covered by Each Service
| Preventive service | Eligible population | Preventive ratio, mean (SD) |
||||
|---|---|---|---|---|---|---|
| No MM | Physical health only | Mental health only | Physical and mental health | All | ||
| Combined preventive index | 301,665 | 33.9 (17.3) | 46.3 (19.3) | 37.5 (17.5) | 44.9 (19.2) | 39.6 (19.2) |
| Cardiovascular screening | ||||||
| Blood pressure screening | 301,665 | 82.6 (26.9) | 83.8 (23.7) | 87.2 (23.5) | 84.2 (23.8) | 83.5 (25.4) |
| Diabetes screening | 152,791 | 36.6 (33.8) | 64.7 (33.8) | 33.7 (31.7) | 58.2 (34.7) | 51.7 (36.2) |
| Lipid screening | 175,454 | 54.3 (45.2) | 86.2 (30.9) | 51.1 (45.3) | 82.2 (34.1) | 73.5 (40.2) |
| Cancer screening | ||||||
| Cervical cancer screening | 156,065 | 52.5 (40.5) | 54.0 (41.4) | 50.7 (40.1) | 52.4 (41.0) | 52.5 (40.7) |
| Breast cancer screening | 74,583 | 32.5 (37.0) | 46.3 (38.7) | 28.9 (35.1) | 41.8 (37.6) | 40.1 (38.1) |
| Colorectal cancer screening | 131,996 | 26.4 (36.9) | 42.5 (41.6) | 23.3 (35.6) | 39.4 (41.5) | 36.1 (40.7) |
| Infectious disease screening | ||||||
| HIV screening | 268,970 | 17.7 (37.1) | 8.6 (27.4) | 20.8 (39.4) | 14.2 (34.0) | 15.5 (35.2) |
| Hepatitis C screening | 107,959 | 4.7 (19.2) | 6.2 (22.2) | 10.9 (29.2) | 12.3 (30.9) | 8.4 (25.9) |
| Chlamydia screening | 25,974 | 31.0 (33.8) | 28.6 (33.6) | 28.9 (34.0) | 29.4 (34.1) | 30.6 (33.9) |
| Influenza vaccine | 301,665 | 12.6 (22.4) | 32.2 (34.5) | 13.9 (23.0) | 29.2 (33.2) | 21.3 (29.8) |
| Pneumonia vaccine | 54,669 | 31.3 (41.9) | 54.1 (44.9) | 24.6 (39.4) | 50.1 (45.0) | 47.5 (45.2) |
| Other services | ||||||
| Substance screening | 301,665 | 21.6 (26.8) | 34.8 (32.0) | 27.6 (28.5) | 36.2 (32.0) | 28.7 (30.3) |
| Depression screening | 301,665 | 17.4 (21.7) | 24.8 (24.1) | 29.5 (27.1) | 33.3 (28.4) | 24.1 (25.5) |
| Abdominal aortic aneurysm screening | 11,518 | 8.9 (25.4) | 15.7 (33.3) | 6.2 (22.6) | 16.5 (33.9) | 14.5 (32.1) |
Note: Preventive index and preventive ratios were defined at a patient-level, covering the entire study period. Preventive ratio estimates the proportion of time a patient was up-to-date for a given preventive service. Preventive ratio is the total person-months covered in a given time period divided by the total person-months eligible for a particular service (e.g., if a patient was eligible for blood pressure screening all 48 months, and was covered for it 12 of those months, they would have been covered for 25% of the time [preventive ratio=25%]). Averaging all patient preventive ratios give the mean preventive ratio for each service. The preventive ratio ranges from 0‒100%.
MM, multimorbidity.
Patients with a “physical only” pattern had higher preventive ratios for diabetes, lipid, and cancer screenings, as well as influenza and pneumonia vaccinations relative to other groups. Patients with a “physical and mental health” pattern had higher preventive ratios for substance use, depression, hepatitis C, and abdominal aortic aneurysm screening compared with other groups.
After adjusting for covariates, there was no significant difference between “no multimorbidity” and “mental health only” groups in receipt of preventive care overall; however, people with physical comorbidities were more up-to-date with overall preventive care, compared with former groups (Table 3). The association between preventive ratios and multimorbidity patterns varied across preventive services, though the magnitudes of associations were small. Among individuals in the “mental health only” group, there was 12% less person-time covered for diabetes screening (rate ratio=0.88, 95% CI=0.85, 0.92), 3% less for blood pressure screening (rate ratio=0.97, 95% CI=0.96, 0.98), and 9% less for lipid screening (rate ratio=0.91, 95% CI=0.87, 0.95) compared with the “no multimorbidity” group. Additionally, individuals in the “mental health only” group had between 7% and 31% less person-time covered compared with the “physical health only” group for the same services (Appendix 3), and between 3% and 5% less compared to “no chronic condition” group, except a non-significant difference for lipid screening (Appendix Table 5). Patients in “mental health only” group were less likely to be up-to date with cervical, breast, and colorectal cancer screenings, chlamydia screening (rate ratio=0.91, 95% CI=0.85, 0.98), and pneumonia vaccination (rate ratio=0.79, 95% CI=0.65, 0.96) compared with the “no multimorbidity” group. For HIV, hepatitis C, and depression screenings, the “mental health only” group was more up-to-date than the “no multimorbidity” group, whereas there were no observed differences for influenza vaccination, substance use, and abdominal aortic aneurysm screenings. Individuals in the “physical only” group were less likely to receive HIV (rate ratio=0.72, 95% CI=0.64, 0.80), chlamydia (rate ratio=0.79, 95% CI=0.72, 0.87), and cervical cancer (rate ratio=0.91, 95% CI=0.89, 0.94) screenings than the “no multimorbidity” group. There was no observed difference for substance use and depression screenings for the “physical only” group compared to the no multimorbidity group.
Table 3.
Association Between Multimorbidity Patterns and Receipt of Preventive Services Comparing to No Multimorbidity Group: Multivariate Regression Models
| Preventive service | No multimorbidity | Physical health only | Mental health only | Physical and mental health |
|---|---|---|---|---|
| ARR (95% CI) | ARR (95% CI) | ARR (95% CI) | ||
| Combined preventive index | ref | 1.11*** (1.09, 1.13) | 1.00 (0.98, 1.01) | 1.06*** (1.05, 1.08) |
| Cardiovascular screening | ||||
| Blood pressure screening | ref | 1.04*** (1.03, 1.05) | 0.97*** (0.96, 0.98) | 0.98*** (0.97, 0.99) |
| Diabetes screening | ref | 1.28*** (1.25, 1.31) | 0.88*** (0.85, 0.92) | 1.18*** (1.15, 1.21) |
| Lipid screening | ref | 1.27*** (1.24, 1.30) | 0.91*** (0.87, 0.95) | 1.26*** (1.23, 1.29) |
| Cancer screening | ||||
| Cervical cancer screening | ref | 0.91*** (0.89, 0.94) | 0.92*** (0.89, 0.96) | 0.92*** (0.89, 0.94) |
| Breast cancer screening | ref | 1.02 (0.97,1.08) | 0.81*** (0.73, 0.90) | 0.96** (0.92, 1.00) |
| Colorectal cancer screening | ref | 1.12*** (1.07, 1.18) | 0.87*** (0.80, 0.95) | 1.15*** (1.09, 1.21) |
| Infectious disease screening | ||||
| HIV screening | ref | 0.72*** (0.64, 0.80) | 1.16*** (1.05, 1.29) | 1.04 (0.92, 1.16) |
| Hepatitis C screening | ref | 1.20*** (1.05, 1.36) | 1.54*** (1.27, 1.86) | 1.70*** (1.49, 1.95) |
| Chlamydia screening | ref | 0.79*** (0.72, 0.87) | 0.91** (0.85, 0.98) | 0.85*** (0.81, 0.90) |
| Influenza vaccine | ref | 1.28*** (1.22, 1.34) | 1.01 (0.96, 1.06) | 1.30*** (1.24, 1.35) |
| Pneumonia vaccine | ref | 1.18*** (1.13, 1.24) | 0.79** (0.65, 0.96) | 1.20*** (1.14, 1.27) |
| Other services | ||||
| Substance screening | ref | 0.97 (0.91, 1.03) | 1.02 (0.96, 1.08) | 0.99 (0.93, 1.05) |
| Depression screening | ref | 0.97 (0.92, 1.02) | 1.33*** (1.26, 1.41) | 1.20*** (1.14, 1.26) |
| Abdominal aortic aneurysm screening | ref | 1.15 (0.93, 1.42) | 1.13 (0.39, 3.33) | 1.31** (1.06, 1.63) |
Note: Boldface indicates statistical significance (*p<0.10; **p<0.05; ***p<0.01). Negative binomial models were conducted to assess the association between type of multimorbidity patterns and preventive ratios for each service separately. The regression models were adjusted for: age, gender, race/ethnicity, primary language, insurance type, rural location, smoking status at the beginning of the study, number of office visits during study period, presence of established primary care provider at the beginning of the study, and usual provider continuity index. Results are presented as an exponentiated β-coefficients or rate ratios (RR) and 95% CI.
ARR, adjusted rate ratio
In sensitivity analyses, regression analyses stratified by age group revealed similar findings (Appendix Table 4).
DISCUSSION
Overall, preventive ratios were low in groups with and without multimorbidity for multiple preventive services. Results from this study are similar to clinic-level analyses, which also showed low rates of preventive care.25 Patients with physical health conditions were more likely to be up-to-date on preventive care than those with “no multimorbidity” or those with “mental health conditions only”. In fact, those with no multimorbidity had the lowest overall preventive index and those in the “mental health only” group were less likely to be up-to-date on a number of preventive ratios. The findings show the value of using preventive ratios over combined a preventive index when comparing these patient populations, as the ratios provided more information about which specific screenings are lacking for each population group.
Although there was no significant difference between “no multimorbidity” and “mental health only” groups for combined preventive index, using preventive ratios allowed the authors to identify subgroups of patients with lower rates of time covered for certain services. Individuals with “mental health only” multimorbidity were less likely to be up-to-date with cardiometabolic and cancer screenings compared with the “no multimorbidity” group or groups with physical conditions. Previous studies have shown that patients with mental health conditions receive fewer preventive care services.12,19,41,42 One potential explanation might be that competing demands for individuals with mental health diagnoses such as psychiatric treatment may become a priority during time-limited office visit, resulting in delayed delivery of preventive care. However, other explanations may exist: For instance, mental illness can be a barrier to care when patients decline preventive care or are not in a position to follow through with advised care. The fragmentation of care in the U.S. further disadvantages populations with mental health conditions who have been shown to receive lower overall quality of care.43 This finding is especially concerning given that some patients with mental health conditions receive antipsychotic medications, which are associated with higher rates of cardiovascular and metabolic disorders; thus, timely preventive screening for cardiovascular and metabolic disorders is critical in these patients.44,45
Interestingly, findings highlighted that the presence of physical conditions was associated with higher rates of preventive care among patients with co-occurring mental health conditions. Individuals in the “physical and mental health” group were more up-to-date with preventive care than the “mental health only” group, though not as up-to-date as the “physical health only” group, despite individuals in the “physical and mental health” group having, on average, more office visits than other groups. Though individuals with physical conditions were older and had more office visits than the “mental health only” group, similar findings were observed in age-stratified sensitivity analysis, and after controlling for number of visits (Appendix Table 4). Patients with physical conditions, whether or not they have mental health disorders, may require more monitoring to manage their conditions, especially cardiovascular disease, which likely aids in receiving guideline-concordant services.
There was no difference with regards to depression screening between the “physical health only” and “no multimorbidity” groups, though the preventive ratio were relatively low. Multimorbidity may lead to the development of mental health disorders such as anxiety and depression.46 Timely screening and treatment for mental health disorders among individuals with physical multimorbidity could improve management of complex multimorbidity cases.
In addition to the explanations of fragmented services and competing demands, a recent qualitative study reported that there was a misconception and negative bias among providers regarding delivery of preventive care to patients with serious mental health conditions.47 Findings from this study and other similar studies suggest the need for interventions to improve preventive care for all patients, especially those with mental health diagnoses. Patients with mental health conditions may benefit from embedding behavioral health providers into primary care settings48 or embedding preventive care delivery at mental health clinics.48,49 Primary care providers might also benefit from additional education about preventive services among patients with mental health comorbidities, and the potential for implicit bias.
This study used EHR data and included a large sample of CHC patients to evaluate the proportion of time up-to-date on various preventive services over 4 years. This approach allowed the authors to capture receipt of preventive care in a consistent manner, especially for services provided on an annual basis. This study included a large percentage of uninsured patients and patients with a significant chronic condition burden, especially with higher mental health and substance abuse burden, who often are less well represented in the other studies.
Limitations
Although use of EHR data had many advantages, such as avoiding recall bias, it may also have some limitations in capturing enrollment continuity within CHCs such as preventive care received at different sites or missing historical data on preventive care. However, studies suggest overall good agreement between EHR data and other sources50 and that CHC patients, especially with chronic conditions, are more likely to remain in CHCs than patients from other settings.51
Individuals with mental health conditions may seek services in community mental health clinics, where they may receive some preventive care potentially underestimating the preventive ratios. Patients with less frequent visits may have undiagnosed conditions, thus leading to underestimation of the proportion of patients with multimorbidity. These results maybe not be generalizable to all CHC patients or to patients receiving care outside CHCs. Nonetheless, previous evidence comparing delivery of preventive care in CHCs and Kaiser Permanente found a similar association with mental health conditions and preventive care delivery in both care settings.20 The authors were not able to control for individual preferences and risk factors, patients’ education level, patient–provider relationships, severity of chronic conditions, or site-level characteristics (e.g., ZIP code–level poverty rates), which may affect access to preventive care. Future studies are needed to further evaluate how unique combinations of chronic conditions affect delivery of preventive care.
CONCLUSIONS
This study provides critical insights into receipt of preventive service among adults with multimorbidity using a more precise method for measuring up-to-date preventive care delivery. There was a variation in receipt of preventive care between multimorbidity groups: Individuals with “mental health only” multimorbidity were less likely to be up-to-date with cardiometabolic and cancer screenings compared with the “no multimorbidity” group or groups with physical conditions, and the “physical only” multimorbidity group had low rates of depression screening as compared with other groups. This study identified vulnerable populations for future interventions to improve preventive care delivery.
Supplementary Material
ACKNOWLEDGMENTS
This work was supported by the National Cancer Institute grant number R01CA181452, and the National Institute On Aging of the National Institutes of Health grant number R01AG061386. The authors would like to acknowledge the participation of our partnering health systems and OCHIN, Inc.
Footnotes
The views presented in this article are solely the responsibility of the authors and do not necessarily represent the views of the funding agencies.
No financial disclosures were reported by the authors of this paper.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES
- 1.Maciosek MV, Coffield AB, Flottemesch TJ, Edwards NM, Solberg LI. Greater use of preventive services in U.S. health care could save lives at little or no cost. Health Aff (Millwood). 2010;29(9):1656–1660. 10.1377/hlthaff.2008.0701. [DOI] [PubMed] [Google Scholar]
- 2.Borsky A, Zhan C, Miller T, Ngo-Metzger Q, Bierman AS, Meyers D. Few Americans receive all high-priority, appropriate clinical preventive services. Health Aff (Millwood). 2018;37(6):925–928. 10.1377/hlthaff.2017.1248. [DOI] [PubMed] [Google Scholar]
- 3.Ward BW, Schiller JS, Goodman RA. Multiple chronic conditions among US adults: a 2012 update. Prev Chronic Dis. 2014;11:E62 10.5888/pcd11.130389. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Yarnall KS, Pollak KI, Ostbye T, Krause KM, Michener JL. Primary care: is there enough time for prevention? Am J Public Health. 2003;93(4):635–641. 10.2105/ajph.93.4.635. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Fenton JJ, Von Korff M, Lin EH, Ciechanowski P, Young BA. Quality of preventive care for diabetes: effects of visit frequency and competing demands. Ann Fam Med. 2006;4(1):32–39. 10.1370/afm.421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Korownyk C, McCormack J, Kolber MR, Garrison S, Allan GM. Competing demands and opportunities in primary care. Can Fam Physician. 2017;63(9):664–668. [PMC free article] [PubMed] [Google Scholar]
- 7.Turner BJ, Hollenbeak CS, Weiner M, Ten Have T, Tang SS. Effect of unrelated comorbid conditions on hypertension management. Ann Intern Med. 2008;148(8):578–586. 10.7326/0003-4819-148-8-200804150-00002. [DOI] [PubMed] [Google Scholar]
- 8.Owens MD, Beckles GL, Ho KK, Gorrell P, Brady J, Kaftarian JS. Women with diagnosed diabetes across the life stages: underuse of recommended preventive care services. J Womens Health (Larchmt). 2008;17(9):1415–1423. 10.1089/jwh.2008.1125. [DOI] [PubMed] [Google Scholar]
- 9.Higashi T, Wenger NS, Adams JL, et al. Relationship between number of medical conditions and quality of care. N Engl J Med. 2007;356(24):2496–2504. 10.1056/nejmsa066253. [DOI] [PubMed] [Google Scholar]
- 10.Bili A, Schroeder LL, Ledwich LJ, Kirchner HL, Newman ED, Wasko MC. Patterns of preventive health services in rheumatoid arthritis patients compared to a primary care patient population. Rheumatol Int. 2011;31(9):1159–1165. 10.1007/s00296-010-1461-y. [DOI] [PubMed] [Google Scholar]
- 11.Harrison SM, Wei MY, Lamerato LE, Petrie JG, Toth Martin E. Multimorbidity is associated with uptake of influenza vaccination. Vaccine. 2018;36(25):3635–3640. 10.1016/j.vaccine.2018.05.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Druss BG, Rosenheck RA, Desai MM, Perlin JB. Quality of preventive medical care for patients with mental disorders. Med Care. 2002;40(2):129–136. 10.1097/00005650-200202000-00007. [DOI] [PubMed] [Google Scholar]
- 13.Katon W, Lin EH, Kroenke K. The association of depression and anxiety with medical symptom burden in patients with chronic medical illness. Gen Hosp Psychiatry. 2007;29(2):147–155. 10.1016/j.genhosppsych.2006.11.005. [DOI] [PubMed] [Google Scholar]
- 14.Institute of Medicine Committee on Assuring the Health of the Public in the 21st Century. The Future of the Public’s Health in the 21st Century. Washington, DC: National Academies Press (US), 2002. 10.17226/10548. [DOI] [Google Scholar]
- 15.Walker ER, McGee RE, Druss BG. Mortality in mental disorders and global disease burden implications: a systematic review and meta-analysis. JAMA Psychiatry. 2015;72(4):334–341. 10.1001/jamapsychiatry.2014.2502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Druss BG, Zhao L, Von Esenwein S, Morrato EH, Marcus SC. Understanding excess mortality in persons with mental illness: 17-year follow up of a nationally representative US survey. Med Care. 2011;49(6):599–604. 10.1097/mlr.0b013e31820bf86e. [DOI] [PubMed] [Google Scholar]
- 17.Baller JB, McGinty EE, Azrin ST, Juliano-Bult D, Daumit GL. Screening for cardiovascular risk factors in adults with serious mental illness: a review of the evidence. BMC Psychiatry. 2015;15:55 10.1186/s12888-015-0416-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Lawrence D, Kisely S. Review: inequalities in healthcare provision for people with severe mental illness. J Psychopharmacol. 2010;24(4 suppl):61–68. 10.1177/1359786810382058. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Streit S, da Costa BR, Bauer DC, et al. Multimorbidity and quality of preventive care in Swiss university primary care cohorts. PLoS One. 2014;9(4):e96142 10.1371/journal.pone.0096142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yarborough BJH, Perrin NA, Stumbo SP, Muench J, Green CA. Preventive service use among people with and without serious mental illnesses. Am J Prev Med. 2018;54(1):1–9. 10.1016/j.amepre.2017.08.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mair FS, May CR. Thinking about the burden of treatment. BMJ. 2014;349:g6680 10.1136/bmj.g6680. [DOI] [PubMed] [Google Scholar]
- 22.Spencer-Bonilla G, Quinones AR, Montori VM. Assessing the burden of treatment. J Gen Intern Med. 2017;32(10):1141–1145. 10.1007/s11606-017-4117-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Fried TR, Tinetti ME, Iannone L. Primary care clinicians’ experiences with treatment decision making for older persons with multiple conditions. Arch Intern Med. 2011;171(1):75–80. 10.1001/archinternmed.2010.318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Vogt TM, Aickin M, Ahmed F, Schmidt M. The prevention index: using technology to improve quality assessment. Health Serv Res. 2004;39(3):511–530. 10.1111/j.1475-6773.2004.00242.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hatch BA, Tillotson CJ, Huguet N, Hoopes MJ, Marino M, DeVoe JE. Use of a preventive index to examine clinic-level factors associated with delivery of preventive care. Am J Prev Med. 2019;57(2):241–249. 10.1016/j.amepre.2019.03.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Akinyemiju T, Jha M, Moore JX, Pisu M. Disparities in the prevalence of comorbidities among US adults by state Medicaid expansion status. Prev Med. 2016;88:196–202. 10.1016/j.ypmed.2016.04.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Shin P, Alvarez C, Sharac J, et al. A Profile of Community Health Center Patients: Implications for Policy. The Henry J. Kaiser Family Foundation, 2013. [Google Scholar]
- 28.Devoe JE, Gold R, Spofford M, et al. Developing a network of community health centers with a common electronic health record: description of the Safety Net West Practice-based Research Network (SNW-PBRN). J Am Board Fam Med. 2011;24(5):597–604. 10.3122/jabfm.2011.05.110052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Devoe JE, Sears A. The OCHIN community information network: bringing together community health centers, information technology, and data to support a patient-centered medical village. J Am Board Fam Med. 2013;26(3):271–278. 10.3122/jabfm.2013.03.120234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wallace J, Sommers BD. Health insurance effects on preventive care and health: a methodologic review. Am J Prev Med. 2016;50(5 suppl 1):S27–S33. 10.1016/j.amepre.2016.01.003. [DOI] [PubMed] [Google Scholar]
- 31.USPSTF A and B Recommendations. U.S. Preventive Services Task Force; www.uspreventiveservicestaskforce.org/Page/Name/uspstf-a-and-b-recommendations/ Published September 2019. Accessed May 11, 2020. [Google Scholar]
- 32.CDC Recommended adult immunization schedule for ages 19 years or older. www.cdc.gov/vaccines/schedules/downloads/adult/adult-combined-schedule.pdf Published 2017. Accessed May 11, 2020.
- 33.CMS Chronic Conditions Data Warehouse (CCW). www.ccwdata.org Accessed April 5, 2020.
- 34.Prados-Torres A, Poblador-Plou B, Calderon-Larranaga A, et al. Multimorbidity patterns in primary care: interactions among chronic diseases using factor analysis. PLoS One. 2012;7(2):e32190 10.1371/journal.pone.0032190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Finkelstein MM. Preventive screening. What factors influence testing? Can Fam Physician. 2002;48:1494–1501. [PMC free article] [PubMed] [Google Scholar]
- 36.DeVoe JE, Fryer GE, Phillips R, Green L. Receipt of preventive care among adults: insurance status and usual source of care. Am J Public Health. 2003;93(5):786–791. 10.2105/ajph.93.5.786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Mainous AG 3rd, Koopman RJ, Gill JM, Baker R, Pearson WS. Relationship between continuity of care and diabetes control: evidence from the Third National Health and Nutrition Examination Survey. Am J Public Health. 2004;94(1):66–70. 10.2105/ajph.94.1.66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Breslau N, Haug MR. Service delivery structure and continuity of care: a case study of a pediatric practice in process of reorganization. J Health Soc Behav. 1976;17(4):339–352. 10.2307/2136712. [DOI] [PubMed] [Google Scholar]
- 39.Pham HH, Schrag D, Hargraves JL, Bach PB. Delivery of preventive services to older adults by primary care physicians. JAMA. 2005;294(4):473–481. 10.1001/jama.294.4.473. [DOI] [PubMed] [Google Scholar]
- 40.Nicholson K, Terry AL, Fortin M, Williamson T, Bauer M, Thind A. Prevalence, characteristics, and patterns of patients with multimorbidity in primary care: a retrospective cohort analysis in Canada. Br J Gen Pract. 2019;69(686):e647–e656. . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Werneke U, Horn O, Maryon-Davis A, Wessely S, Donnan S, McPherson K. Uptake of screening for breast cancer in patients with mental health problems. J Epidemiol Community Health. 2006;60(7):600–605. 10.1136/jech.2005.039065. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Xiong GL, Bermudes RA, Torres SN, Hales RE. Use of cancer-screening services among persons with serious mental illness in Sacramento County. Psychiatr Serv. 2008;59(8):929–932. 10.1176/ps.2008.59.8.929. [DOI] [PubMed] [Google Scholar]
- 43.Morden NE, Mistler LA, Weeks WB, Bartels SJ. Health care for patients with serious mental illness: family medicine’s role. J Am Board Fam Med. 2009;22(2):187–195. 10.3122/jabfm.2009.02.080059. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Correll CU, Solmi M, Veronese N, et al. Prevalence, incidence and mortality from cardiovascular disease in patients with pooled and specific severe mental illness: a large-scale meta-analysis of 3,211,768 patients and 113,383,368 controls. World Psychiatry. 2017;16(2):163–180. 10.1002/wps.20420. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.American Diabetes Association American Psychiatric Association, American Association of Clinical Endocrinologists, North American Association for the Study of Obesity. Consensus development conference on antipsychotic drugs and obesity and diabetes. J Clin Psychiatry. 2004;65(2):267–272. 10.4088/jcp.v65n0219. [DOI] [PubMed] [Google Scholar]
- 46.Gould CE, O’Hara R, Goldstein MK, Beaudreau SA. Multimorbidity is associated with anxiety in older adults in the Health and Retirement Study. Int J Geriatr Psychiatry. 2016;31(10):1105–1115. 10.1002/gps.4532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Stumbo SP, Yarborough BJH, Yarborough MT, Green CA. Perspectives on providing and receiving preventive health care from primary care providers and their patients with mental illnesses. Am J Health Promot. 2018;32(8):1730–1739. 10.1177/0890117118763233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Scharf DM, Schmidt Hackbarth N, Eberhart NK, et al. General medical outcomes from the primary and behavioral health care integration grant program. Psychiatr Serv. 2016;67(11):1226–1232. 10.1176/appi.ps.201500352. [DOI] [PubMed] [Google Scholar]
- 49.Druss BG, von Esenwein SA, Glick GE, et al. Randomized trial of an integrated behavioral health home: the Health Outcomes Management and Evaluation (HOME) Study. Am J Psychiatry. 2017;174(3):246–255. 10.1176/appi.ajp.2016.16050507. [DOI] [PubMed] [Google Scholar]
- 50.Bailey SR, Heintzman JD, Marino M, et al. Measuring preventive care delivery: comparing rates across three data sources. Am J Prev Med. 2016;51(5):752–761. 10.1016/j.amepre.2016.07.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Huguet N, Kaufmann J, O’Malley J, et al. Using electronic health records to evaluate natural experiments: accounting for patient attrition. Med Care. In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.