Sage Journals: Discover world-class research

Abstract

Objective:

Preexposure Prophylaxis (PrEP) is under-utilized in primary care. Given differences in treatment approaches for other conditions between family medicine (FM) and general internal medicine (GIM), this study compared PrEP-prescribing between FM and GIM physicians.

Methods:

De-identified electronic health record data from a multi-state health care system was used in this retrospective observational study. The time period from 1/1/13 to 9/30/21 was used to identify PrEP eligible patients using measures of current sexually transmitted disease and condomless sex at the time of eligibility. Receipt of PrEP was measured in the 12 months after PrEP eligibility. The odds of receiving PrEP in GIM as compared to FM was computed before and after adjusting for demographics and physical and psychiatric comorbidities.

Results:

The majority of eligible patients were 18 to 39 years of age, 60.9% were female and 71.6% were White race. Among PrEP eligible patients, 1.1% received PrEP in the first year after index date. Receiving PrEP was significantly more likely among patients treated in GIM versus FM (OR = 2.30; 95% CI:1.63-3.25). After adjusting for covariates, this association remained statistically significant (OR = 2.02; 95% CI:1.41-2.89).

Conclusions:

PrEP is grossly under-utilized in primary care. The majority of Americans enter the health care system through primary care and not through HIV providers or other specialties. Therefore, educational interventions are needed to increase confidence and knowledge and to encourage PrEP prescribing by FM and GIM physicians.

Keywords

cohort health services research preexposure prophylaxis PrEP primary care prevention

Introduction

Preexposure Prophylaxis (PrEP) is an effective means to reduce transmission of HIV. Over the past decade, use of PrEP has increased. In 2015, just 3% of people who could benefit received PrEP, but this rose to 25% in 2020.¹ Given the potential to prevent HIV, under-utilization of PrEP remains a major public health concern, particularly in primary care where prescribing rates remain low. Barriers to starting PrEP include provider level factors. Physician knowledge, comfort, and willingness to prescribe PrEP can be barriers to offering treatment to eligible patients.² Providers prescribe PrEP more frequently if they treat patients with HIV and if they report confidence and knowledge about PrEP.² HIV providers are more PrEP-knowledgeable than primary care physicians, but there is evidence that HIV specialists do not perceive this as their responsibility and primary care providers report lack of knowledge as a barrier to being the main source of PrEP treatment.^2
-4 Nearly two-thirds of PrEP prescribers are family medicine (FM) or general internal medicine (GIM) physicians⁵ which is consistent with evidence from a qualitative study that indicated a majority of primary care providers agree that PrEP should be offered in primary care.⁶ However, providers less comfortable with sexual history taking are less likely to prescribe PrEP,⁶ and this discomfort is more common in primary care than in HIV specialty care.⁴

While it is logical that HIV specialists, as compared to primary care physicians, have more confidence and knowledge about PrEP and are more likely to prescribe PrEP, it is not known whether PrEP prescribing differs within primary care specialties. Prior studies have shown variation between FM and GIM providers in confidence related to treating different conditions.⁷ For example, there is evidence that self-reported preparedness to treat numerous common conditions differs between FM and GIM providers.⁷ A significantly higher percent of GIM as compared to FM residents self-reported being very prepared to diagnose and treat diabetes, hyperlipidemia and hypertension and more FM as compared to GIM residents were very prepared to diagnose and treat depression, low back pain, upper respiratory tract infections, and vaginitis.⁷ There are significant differences in the types of screening, testing and procedures performed by General Practice/FM physicians compared to GIM physicians.⁸ Patients with anxiety and depression are more likely to receive an antidepressant if treated in FM as compared to GIM.⁹ Sexual history taking for adolescent females is significantly less likely among internal medicine compared to FM providers.¹⁰ Such practice differences, particularly around sexual history taking and stigmatized conditions such as depression, suggests there may also be differences between FM and GIM physicians in discussing sexual history and prescribing PrEP.

Although two-thirds of PrEP prescriptions are written in primary care,⁵ PrEP initiation in FM and GIM is low relative to specialty care. Therefore, determining whether 1 primary care discipline or both have low prescribing rates is important to inform educational and training interventions. Additionally, identifying the magnitude of missed PrEP initiation opportunities in FM and GIM is a major public health concern. We are not aware of any research comparing PrEP prescribing between FM and GIM physicians. Nonetheless, practice differences between FM and GIM exist for other medical diagnoses, and it is possible that PrEP prescribing rates may differ between FM and GIM physicians. The present study was designed to determine if the odds of receiving a PrEP prescription among eligible patients differed between FM and GIM providers.

Methods

Data Source

De-identified electronic medical record data were obtained from the Saint Louis University-SSM (SLU-SSM) Healthcare System’s Virtual Data Warehouse (VDW). Because SLU-SSM is a member site in the Health Care Systems Research Network (HCSRN) (www.hcsrn.org), the VDW was created according to HCSRN specifications. The SLU-SSM VDW includes approximately 5 million patients from birth to >90 years of age who had ambulatory and hospital encounters at academic and non-academic SSM health care locations since 1/1/2008. The database is updated monthly. The multi-state SSM health care system includes rural and urban settings from the St. Louis, Missouri metropolitan area, mid-Missouri, southern Illinois, Oklahoma City and surrounding metropolitan area, and southern Wisconsin.

VDW variables were created from ICD-9 and ICD-10 diagnostic codes, Current Procedural Terminology (CPT) codes, pharmacy orders, laboratory orders and results, vital signs, provider and clinic type, and demographics. The Saint Louis University Institutional Review Board reviewed research using the VDW as exempt because all data is historical and de-identified. Previous studies using the VDW have been published.¹¹

Eligibility Criteria

We followed a method described by Caponi et al¹² to identify patients eligible for PrEP based on diagnoses of sexually transmitted infection or an ICD-9 or ICD-10 codes for condomless sex and other sexual behavior that could increase HIV risk. Specifically, PrEP eligible patients had 1 or more diagnosis between 1/1/13 to 9/30/21 for the following: gonorrhea, syphilis, chlamydia, trichomoniasis, and/or codes for conditions affecting health which included contact/suspected exposure to infections with predominately sexual mode of transmission, contact with and suspected exposure to HIV, and/or ICD-9 or ICD-10 codes for condomless sex, high risk, and other sexual behavior that could increase HIV risk.¹² This time frame was chosen because 1) the first drug for PrEP was approved in July 2012 (Truvada) and 2) the time period allowed for 1 year to search for outcomes, that is, a PrEP order.

The index date was the date of the first PrEP eligible diagnosis occurring in either FM or GIM clinics. Patients were ≥18 years of age at index date. To ensure the sample consisted of established patients and not new patients, we required at least 1 visit in the 2 years prior and in the year after index to the same department (FM or GIM) in which PrEP eligible diagnoses occurred. For example, we considered a patient with a syphilis diagnosis in FM or GIM to be PrEP eligible for the 12-month period following diagnosis. We used a rolling enrollment so patients entered the study the first time an eligible diagnosis occurred in 2013 to 2021 and this initial diagnosis must have occurred in FM or GIM.

We excluded patients with contraindications for PrEP which included diagnoses for HIV (n = 188), hepatitis B (n = 80) or chronic kidney disease (n = 676). Patients with prevalent PrEP in the 2-years prior to index were excluded. This left 14 252 patients meeting criteria for PrEP eligibility in either FM or GIM at index date. A diagram of the sampling approach is shown in Figure 1.

Figure 1.

Sampling diagram.

Exposure Variable

FM versus GIM clinic where a patient was first eligible for PrEP.

Outcome

The outcome was a prescription for Descovy (emtricitabine/tenofovir alafenamide; approved 10/3/19) or Truvada (emtricitabine/tenofovir disoproxil fumarate; approved 7/16/12) within a year after index date. The prescription must have been from FM or GIM and from the same department that made the diagnoses we used to define patients who were PrEP eligible.

Covariates

Covariates included year of enrollment or eligibility for PrEP, age, sex, race, and comorbid conditions (measured in the 2-years prior to index). Sex (male or female) and race are obtained when patients register for care and is considered “legal” sex in many states. Covariates were selected because they have been associated with odds of starting PrEP.¹³ The following age categories were modeled: 18 to 29, 30 to 39, 40 to 49, 50 to 59, and ≥60 years of age. Race included White, Black, Other, and unknown. Physical comorbidities were measured using the Charlson Comorbidity Index (CCI).^14,15 Higher CCI scores indicate greater morbidity and risk for mortality and those with higher scores are likely receiving more health care which could contribute to detection bias. In addition, older patients, more prevalent in GIM, should have higher CCI scores than patients seen in FCM. Psychiatric comorbidities included depression, anxiety disorder, severe mental illness (ie, schizophrenia and/or bipolar disorder), any form of substance abuse/dependence, and nicotine dependence/smoking. Nicotine dependence/smoking is an addictive behavior and a proxy for less than optimal self-care. Detailed diagnostic algorithms and definitions for all variables are shown in Supplemental Appendix A, e-table 1.

Analytic Approach

All analyses were conducted in SAS v9.4 (SAS Institute, Cary, NC) at an alpha of 0.05. Characteristics of patients were summarized as means (±SD) or frequencies and percents. The characteristics of patients in FM versus GIM as well as those receiving versus not receiving PrEP were compared using chi-square tests, Fisher’s exact tests, and independent samples t-tests where appropriate. Standardized mean difference percent (SMD% = SMD*100) was a measure of effect size when calculating bivariate comparisons; an SMD% > 10 is considered meaningfully different.¹⁶ Logistic regression models calculated odds ratios and 95% confidence intervals for the relationship of GIM versus FM and PrEP prescription before and after adjusting for all covariates/comorbidities.

Results

As shown in Table 1, 83% of PrEP eligible patients were first diagnosed in FM and 17% in GIM. Among PrEP eligible patients, 1.1% received PrEP in the first year after index date. Figure 2 shows PrEP prescribing among eligible patients by year of eligibility and that prescribing among eligible patients increased. The majority of patients were 18 to 39 years of age, 60.9% were female and 71.6% were White race. The mean comorbidity index was 0.4 ± 0.9. Nicotine dependence/smoking was the most common comorbid psychiatric disorder. Over 50% of PrEP eligible patients had a diagnosis indicating a suspected exposure to infections with sexual model of transmission, followed by about 20% with a diagnosis of sexual behavior that could increase risk of HIV transmission. Chlamydia was the most common sexually transmitted infection (12.0%).

Table 1.

Characteristics (%) of PrEP Eligible Patients (n = 14 252).

Variable	Overall (n = 14 252)
Department
Family medicine (FM)	11 829 (83.0)
General internal medicine (GIM)	2423 (17.0)
PrEP prescription^a	151 (1.1)
Eligible year
2013	1301 (9.1)
2014	1342 (9.4)
2015	1829 (12.8)
2016	1752 (12.3)
2017	1838 (12.9)
2018	1816 (12.7)
2019	1681 (11.8)
2020	1466 (10.3)
2021	1227 (8.6)
Eligibility diagnoses
Gonorrhea	390 (2.7)
Syphilis	272 (1.9)
Chlamydia	1710 (12.0)
Trichomoniasis	1048 (7.4)
Suspected exposure to infections w/ sexual mode of transmission	7430 (52.1)
Suspected exposure to HIV	1226 (8.6)
High risk sexual behavior	2656 (18.6)
Sociodemographic-related
Age
18-29	6626 (46.5)
30-39	3160 (22.2)
40-49	2244 (15.7)
50-59	1436 (10.1)
≥60	786 (5.5)
Female gender	8682 (60.9)
Race
White	10 204 (71.6)
Black	2987 (21.0)
Other	695 (4.9)
Unknown	366 (2.6)
Comorbidities^b
Charlson index, mean (±SD)	0.4 (±0.9)
Depression	1694 (11.9)
Anxiety^c	2450 (17.2)
Severe mental health^d	707 (5.0)
Substance abuse/dependence^e	1003 (7.0)
Nicotine dependence/smoking	4084 (28.7)

Prep prescription in 1-year after index (had to occur in same department as PrEP eligible diagnosis).

Comorbidities measured from 2-years prior to index to index date.

Anxiety disorders = PTSD, panic disorder, OCD, social phobia, GAD, anxiety NOS.

Severe mental health = bipolar or schizophrenia.

Substance abuse/dependence—drug or alcohol.

Figure 2.

Percent PrEP by eligiblity year and department.

Characteristics of PrEP eligible patients by treatment in FM versus GIM are shown in Table 2. A significantly higher proportion of PrEP eligible patients received a PrEP prescription if treated in GIM as compared to FM (2.0 vs 0.9%; P < .0001; SMD% = 9.4). Younger age, female gender, and white race were all more prevalent in patients treated in FM as compared to GIM (P < .0001; SMD range: 3.6-36.3). The mean Charlson comorbidity index score was significantly less in FM versus GIM (P < .0001; SMD% = 13.8). Depression (P < .0001; SMD% = 7.8), anxiety disorders (P < .0001; SMD% = 10.1) and nicotine dependence/smoking (P = 0.011; SMD% = 5.7) were all significantly more prevalent in FM versus GIM patients. The prevalence of severe mental illness and substance abuse/dependence did not significantly differ in FM versus GIM.

Table 2.

Characteristics (%) of PrEP Eligible Patients by Department (n = 14 252).

Variable	Family medicine (n = 11 829)	General internal medicine (n = 2423)	P-value	SMD%
PrEP prescription^a	103 (0.9)	48 (2.0)	<.0001^f	9.4
Sociodemographic-related
Age			<.0001^f
18-29	5700 (48.2)	926 (38.2)		20.2
30-39	2705 (22.9)	455 (18.8)		10.1
40-49	1826 (15.4)	418 (17.3)		4.9
50-59	1051 (8.9)	385 (15.9)		21.4
≥60	547 (4.6)	239 (9.9)		20.3
Female gender	7563 (63.9)	1119 (46.2)	<.0001^f	36.3
Race			<.0001^f
White	8720 (73.7)	1484 (61.3)		26.9
Black	2256 (19.1)	731 (30.2)		26.0
Other	592 (5.0)	103 (4.3)		3.6
Unknown	261 (2.2)	105 (4.3)		12.0
Comorbidities^b
Charlson index, mean (±SD)	0.3 (±0.8)	0.5 (±1.1)	<.0001^g	13.8
Depression	1455 (12.3)	239 (9.9)	.001^f	7.8
Anxiety^c	2108 (17.8)	342 (14.1)	<.0001^f	10.1
Severe mental health^d	593 (5.0)	114 (4.7)	.524^f	1.4
Substance abuse/dependence^e	846 (7.2)	157 (6.5)	.238^f	2.7
Nicotine dependence/smoking	3441 (29.1)	643 (26.5)	.011^f	5.7

Abbreviations: SMD%, standardized mean difference percent.

Prep prescription in 1-year after index (had to occur in same department as PrEP eligible diagnosis).

Comorbidities measured from 2-years prior to index to index date.

Anxiety disorders = PTSD, panic disorder, OCD, social phobia, GAD, anxiety NOS.

Severe mental health = bipolar or schizophrenia.

Substance abuse/dependence – drug or alcohol.

Chi-square test.

Independent samples t-test.

As shown in Table 3, a greater proportion of patients with versus without a PrEP prescription were White (SMD% = 10.5) and other race (SMD% = 7.6) while a greater proportion of those without PrEP were Black (SMD% = 26.9) and female (SMD% = 61.5). No comorbidities were significantly related to PrEP. Figure 3 shows that among those with PrEP versus without PrEP, a greater proportion had diagnoses for sexual behavior that may increase risk for HIV transmission (eg, condomless sex). Conversely, a greater proportion of those without PrEP versus with PrEP were diagnosed with suspected exposure to infections with sexual model of transmission, trichomoniasis, and chlamydia.

Table 3.

Characteristics (%) of PrEP Eligible Patients by Receipt of PrEP Prescription (n = 14 252).

Variable	No PrEP prescription (n = 14 101)	Yes PrEP prescription^a (n = 151)	P-value	SMD%
Sociodemographic-related
Age			.256^f
18-29	6549 (46.4)	77 (51.0)		9.1
30-39	3125 (22.2)	35 (23.2)		2.4
40-49	2229 (15.8)	15 (9.9)		17.6
50-59	1418 (10.1)	18 (11.9)		6.0
≥60	780 (5.5)	6 (4.0)		7.3
Female gender	8671 (61.5)	11 (7.3)	<.0001^f	139.0
Race			.001^f
White	10 089 (71.6)	115 (76.2)		10.5
Black	2970 (21.1)	17 (11.3)		26.9
Other	685 (4.9)	10 (6.6)		7.6
Unknown	357 (2.5)	9 (6.0)		17.1
Comorbidities^b
Charlson index, mean (±SD)	0.4 (±0.9)	0.3 (±0.7)	.119^h	11.2
Depression	1678 (11.9)	16 (10.6)	.622^g	4.1
Anxiety^c	2418 (17.2)	32 (21.2)	.190^g	10.3
Severe mental health^d	700 (5.0)	7 (4.6)	1.00^f	1.5
Substance abuse/dependence^e	996 (7.1)	7 (4.6)	.335^f	10.4
Nicotine dependence/smoking	4045 (28.7)	39 (25.8)	.440^g	6.4

Abbreviation: SMD%, standardized mean difference percent.

Prep prescription in 1-year after index (had to occur in same department as PrEP eligible diagnosis).

Comorbidities measured from 2-years prior to index to index date.

Anxiety disorders = PTSD, panic disorder, OCD, social phobia, GAD, anxiety NOS.

Severe mental health = bipolar or schizophrenia.

Substance abuse/dependence—drug or alcohol.

Fisher’s exact test.

Chi-square test.

Independent samples t-test.

Figure 3.

Distribution of eligible diagnoses among those with and without PrEP prescription.

As shown in Table 4, prior to adjusting for covariates, the odds of receiving PrEP in the year after index was more than twice as likely among patients treated in GIM versus FM (OR = 2.30; 95% CI:1.63-3.25). After adjusting for all covariates, patients treated in GIM remained more likely to receive PrEP compared to those treated in FM (OR = 2.02; 95% CI:1.41-2.89). Results also showed a 39% increased odds of PrEP prescribing among eligible patients per year (OR = 1.39; 95% CI:1.29-1.50). Compared to patients 18 to 29 years of age, those 40 to 49 years of age were significantly less likely to receive PrEP (OR = 0.54; 95% CI:0.30-0.96). Females were substantially less likely to receive PrEP as compared to males (OR = 0.04; 95% CI:0.02-0.08). Compared to White patients, Black patients were significantly less likely to receive PrEP (OR = 0.54; 95% CI:0.32-0.92). Those with an anxiety disorder were significantly more likely to receive a PrEP prescription as compared to patients without an anxiety disorder (OR = 1.72; 95% CI: 1.10-2.69) and those with a substance use disorder were less likely to receive PrEP (OR = 0.43; 95% CI:0.19-0.95).

Table 4.

Crude and Fully Adjusted Logistic Regression Estimating the Relationship of Department and PrEP Prescription for PrEP Eligible Patients (n = 14 252) in the Year After Index.^a

Variable	Model 1—Unadjusted	Model 2—Adjusted
Variable	OR (95% CI)	AOR (95% CI)
Department
Family medicine	1.00	1.00
General internal medicine	2.30 (1.63–3.25)	2.02 (1.41–2.89)
Eligibility year (2013–2021)		1.39 (1.29–1.50)
Age
18-29		1.00
30-39		0.88 (0.58–1.33)
40-49		0.54 (0.30–0.96)
50-59		0.94 (0.54–1.63)
≥60		0.54 (0.22–1.28)
Female gender		0.04 (0.02–0.08)
Race
White		1.00
Black		0.54 (0.32–0.92)
Other		1.53 (0.78–3.00)
Unknown		1.55 (0.76–3.18)
Charlson index		0.95 (0.75–1.20)
Depression		1.18 (0.66–2.11)
Anxiety		1.72 (1.10–2.69)
Severe mental health		1.48 (0.65–3.34)
Substance abuse/dependence		0.43 (0.19–0.95)
Nicotine dependence/smoking		0.88 (0.60–1.30)

Adjusted model includes all variables.

Discussion

Patients who were PrEP eligible and treated in GIM were twice as likely to receive an order for PrEP compared to patients treated in FM. This is despite 83% of PrEP eligible patients being diagnosed in FM and 17% diagnosed in GIM (see Table 1). Older age, Black race, female gender, and having a substance use disorder were inversely associated with receiving PrEP and having an anxiety disorder was positively associated with receiving PrEP. PrEP prescribing also increased with time. Differences in PrEP prescribing between GIM and FM remained after adjusting for patient level factors associated with PrEP.

Unfortunately, only 0.9% of FM and 2.0% of GIM providers issued a PrEP prescription. PrEP is grossly underutilized in primary care; however, PrEP under prescribing is a national problem and goes beyond primary care. For example, our results are consistent with evidence from a nationally distributed sample, not limited to primary care, in which 1.3% of eligible patients received PrEP.¹³

Patients at risk for HIV are unlikely to first enter the healthcare system through HIV/infectious disease specialists. It is important to increase uptake of PrEP in primary care, which is the entry point to preventive health care for most Americans. Low PrEP prescribing is inconsistent with the central tenet of primary care which is disease prevention. The U.S. Preventive Services Task Force (USPSTF) strongly recommends that clinicians offer PrEP to persons who are vulnerable to HIV acquisition (grade A recommendation).¹⁷ While other preventative services with similarly strong recommendations from the USPSTF (eg, colon cancer screening) are widely ordered by FM and GIM physicians, there is a clear need to improve PrEP prescribing rates in primary care, regardless of specialty.

It is not clear why prescribing rates are so low in primary care and lower in FM compared to GIM. It is possible that infrequent sexual history taking, less knowledge about PrEP and less confidence prescribing PrEP contribute to underutilization. A study of primary care providers and HIV care observed 39% of providers routinely obtained sexual histories.¹⁸ Torkko and colleagues¹⁰ observed that compared to FM physicians, regular sexual history taking was 4-times more likely in obstetrician/gynecology and 61% less likely in GIM. The former finding is consistent with extremely low PrEP prescribing in primary care, but the latter finding appears inconsistent with our results. Yet, sexual history-taking is just 1 component associated with increased PrEP use. Torkko et al¹⁰ also revealed that PrEP prescribing is much more common if the provider’s patient population consists of >50% of men who have sex with men. The practices under study are more heterogeneous which could contribute to low PrEP prescribing.

Addressing the purview paradox is necessary to increase PrEP initiation in primary care. Specifically, HIV specialists are expert in treating HIV while primary care providers are more likely to care for patients who are HIV negative but are eligible for PrEP.¹⁹ Further research is needed to understand barriers to PrEP prescribing that are common and specific to GIM and FM. Our results suggest a need to augment PrEP education and training for both GIM and FM providers. This is particularly urgent for FM given evidence that more patients met our definition of PrEP eligibility in FM than in GIM. Training and education could occur during residency and through continuing medical education. Education should include HIV related training which is positively associated with greater PrEP prescribing among primary care providers.²⁰ Others have suggested making HIV risk assessment standard of care and increasing physician ability to encourage PrEP uptake as key components of an education program.¹⁹

We observed that Black patients were 46% less likely to receive PrEP which is consistent with prior studies demonstrating a race disparity in use and or access to PrEP.^13,21,22 In a nationally distributed cohort, we recently observed Black patients were 59% less likely to receive PrEP as compared to whites.¹³ There are multiple reasons for this race disparity, and most center around knowledge, stigma and trust. For instance, among men who have sex with men, Black, as compared to white men, were less likely to know about PrEP and to discuss PrEP with their physician.²¹ Distrust of medical providers is correlated with lower PrEP utilization among Black women.²³ Among providers who responded to a survey, there was evidence that providers believed Black patients, as compared to whites, were more likely to engage in condomless sex if prescribed PrEP and this was associated with less willingness to prescribe PrEP to Black patients.²² Qualitative research has revealed that stigma around HIV, stigma related to PrEP and sexual behavior, and homophobia/transphobia each contribute to less PrEP utilization.²⁴

We observed that anxiety disorders were associated with a 72% greater odds of receiving PrEP and this is similar to the 67% increased likelihood of starting PrEP among those with an anxiety disorder in a nationally distributed medical record data base.¹³ Although others have found no association between anxiety and willingness to start PrEP,²⁵ we have now shown a positive association between anxiety disorder and PrEP receipt in a national cohort and a multi-state health care system. Anxiety disorders may be related to seeking PrEP due to fear of HIV and low risk taking.

Interestingly, results from the fully adjusted model indicated that patients 40 to 49 years of age were significantly less likely to receive PrEP compared to those 18 to 29 years of age. All other age groups did not significantly differ in odds of PrEP. More than 70% of patients treated in FM were 18 to 39 years of age while only 57% of those treated in GIM were in this age group. The significant association between age and treatment in FM versus GM may have led to detecting a significant association between age and PrEP in the only age group (40-49 years of age) that was similarly distributed between FM and GIM. Previous studies have observed patients over 50 years of age are less likely to use PrEP.^13,26 Though speculative, it is possible that physicians are less likely to discuss PrEP with older patients because they believe they are less likely to engage in condomless anal and vaginal sex or have multiple sexual partners.

Limitations

The sample was majority female which may have led to results that do not reflect the patients at greatest HIV risk which are men who have sex with men. Although sexual orientation is stored in the EHR, about 95% of values were missing. Results may be less generalizable to those at greatest risk for HIV because our definition of PrEP eligibility included chlamydia and trichomoniasis diagnosis in heterosexual women. Given that FM physicians cared for a higher proportion of women in this study, their patient population may have fewer risk factors for HIV acquisition and thus less PrEP prescribing. We may have under-estimated PrEP prescribing because we excluded all patients with CKD, and not just those with more severe CKD for whom PrEP is contraindicated. However, as shown in Figure 1, only 676 patients were lost for this reason. Although our data came from a multi-state health care system, results may not generalize to other geographic regions. Similarly, the majority White patient population could reduce generalizability. Misclassification could bias findings if we classified some patients as not PrEP eligible when they were. Unmeasured confounding could bias results.

Conclusions

PrEP is markedly under-utilized in primary care and particularly in FM. Education and implementation initiatives are needed to increase PrEP prescribing. Research is needed to identify and ameliorate barriers to PrEP in primary care and identify any unique obstacles specific to FM versus GIM.

Supplemental Material

sj-docx-1-jpc-10.1177_21501319231201784 – Supplemental material for Differences Between General Internal Medicine and Family Medicine Physicians’ Initiation of Pre-Exposure Prophylaxis

Supplemental material, sj-docx-1-jpc-10.1177_21501319231201784 for Differences Between General Internal Medicine and Family Medicine Physicians’ Initiation of Pre-Exposure Prophylaxis by Theresa Drallmeier, Joanne Salas, Elizabeth Keegan Garrett, Ashley Meyr, Jane Tucker and Jeffrey F. Scherrer in Journal of Primary Care & Community Health

Footnotes

Author Contributions

1) Substantial contributions to conception and design, acquisition of data or analysis and interpretation of data—study conception by Drallmeier and Scherrer, study design by Drallmeier, Scherrer and Salas, data analyses by Salas. Interpretation of results—all authors.

2) Drafting the article or revising it critically for important intellectual content—all authors.

3) Final approval of the version to be published—all authors.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding to develop and maintain the Virtual Data Warehouse was provided by the Saint Louis University Research Institute. Dr. Scherrer and Ms. Salas had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

ORCID iD

Jeffrey F. Scherrer

Supplemental Material

Supplemental material for this article is available online.

References

Centers for Disease Control and Prevention. PrEP for HIV prevention in the U.S. Accessed August 7, 2023. https://www.cdc.gov/nchhstp/newsroom/fact-sheets/hiv/PrEP-for-hiv-prevention-in-the-US-factsheet.html

Pleuhs

Quinn

Walsh

Petroll

John

SA.

Health care provider barriers to HIV pre-exposure prophylaxis in the United States: a systematic review. AIDS Patient Care STDS. 2020;34(3):111-123.

Agovi

Anikpo

Cvitanovich

Craten

Asuelime

Ojha

RP.

Knowledge needs for implementing HIV pre-exposure prophylaxis among primary care providers in a safety-net health system. Prev Med Rep. 2020;20:101266.

Petroll

Walsh

Owczarzak

McAuliffe

Bogart

Kelly

JA.

PrEP awareness, familiarity, comfort, and prescribing experience among US primary care providers and HIV specialists. AIDS Behav. 2017;21(5):1256-1267.

Chan

Chappel

Maddox

KEJ

, et al. Pre-exposure prophylaxis for preventing acquisition of HIV: a cross-sectional study of patients, prescribers, uptake, and spending in the United States, 2015-2016. PLoS Med. 2020;17(4):e1003072.

Storholm

Ober

Mizel

, et al. Primary care providers’ knowledge, attitudes, and beliefs about HIV pre-exposure prophylaxis (PrEP): informing network-based interventions. AIDS Educ Prev. 2021;33(4):325-344.

Wiest

Ferris

Gokhale

Campbell

Weissman

Blumenthal

Preparedness of internal medicine and family practice residents for treating common conditions. JAMA. 2002;288(20):2609-2614.

Strumann

Flagel

Emcke

Steinhauser

Procedures performed by general practitioners and general internal medicine physicians - a comparison based on routine data from Northern Germany. BMC Fam Pract. 2018;19(1):189.

Brieler

Scherrer

Salas

Differences in prescribing patterns for anxiety and depression between general internal medicine and family medicine. J Affect Disord. 2015;172:153-158.

10.

Torkko

Gershman

Crane

Hamman

Baron

Testing for Chlamydia and sexual history taking in adolescent females: results from a statewide survey of Colorado primary care providers. Pediatrics. 2000;106(3):E32.

11.

Salas

Hinyard

Cappellari

, et al. Infant, pediatric and adult well visit trends before and during the COVID-19 pandemic: a retrospective cohort study. BMC Health Serv Res. 2022;22(1):328.

12.

Caponi

Burgess

Leatherwood

Molano

FL.

Demographic characteristics associated with the use of HIV pre-exposure prophlaxis (PrEP) in an urban, community health center. Prev Med Rep. 2019;15:100889.

13.

Drallmeier

Garrett

Meyr

Salas

Scherrer

JF.

Demographic factors, psychiatric and physical comorbidities associated with starting preexposure prophylaxis in a nationally distributed cohort. Prev Med. 2022;164:107344.

14.

Charlson

Pompei

Ales

MacKenzie

CR.

A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987; 40(5):373-383.

15.

Sharabiani

Aylin

Bottle

Systematic review of comorbidity indices for administrative data. Med Care. 2012;50(12):1109-1118.

16.

Austin

Stuart

EA.

Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med. 2015;34(28):3661-3679.

17.

U.S. Preventive Services Task Force. Prevention of human immunodeficiency virus (HIV) infection: preexposure prophylaxis. Accessed January 10, 2023. https://www.uspreventiveservicestaskforce.org/uspstf/recommendation/prevention-of-human-immunodeficiency-virus-hiv-infection-pre-exposure-prophylaxis

18.

Townes

Gaul

Henny

KD.

Sexual health discussion practices and HIV clinical care provided by primary care providers in the Southeast United States, K-BAP Study (2017-2018). Fam Pract. 2022;40:39–46.

19.

Cooper

Juarez

Morris

, et al. Recommendations for increasing physician provision of pre-exposure prophylaxis: Implications for medical student training. Inquiry. 2021;58:469580211017666.

20.

Henny

Duke

Geter

, et al. HIV-related training and correlates of knowledge, HIV screening and prescribing of nPEP and PrEP among primary care providers in Southeast United States, 2017. AIDS Behav. 2019;23(11):2926-2935.

21.

Kanny

Jeffries

WLt

Chapin-Bardales

, et al. Racial/ethnic disparities in HIV preexposure prophylaxis among men who have sex with men - 23 Urban Areas, 2017. MMWR Morb Mortal Wkly Rep. 2019;68(37):801-806.

22.

Calabrese

SK.

Understanding, contextualizing, and addressing PrEP stigma to enhance PrEP implementation. Curr HIV/AIDS Rep. 2020;17(6):579-588.

23.

Tekeste

Hull

Dovidio

, et al. Differences in medical mistrust between Black and White women: Implications for patient-provider communication about PrEP. AIDS Behav. 2019;23(7):1737-1748.

24.

Kerr

Ayangeakaa

Bullock

NAA

, et al. A qualitative exploration of various stigmas impacting HIV pre-exposure prophylaxis (PrEP) uptake among African American young adults. Fam Community Health. 2022;45(4):218-227.

25.

Miller

Harrison

Sanasi-Bhola

A scoping review investigating relationships between depression, anxiety, and the PrEP care continuum in the United States. Int J Environ Res Public Health. 2021;18(21):11431.

26.

Golub

Fikslin

Goldberg

Pena

Radix

Predictors of PrEP uptake among patients with equivalent access. AIDS Behav. 2019;23(7):1917-1924.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB