Abstract
Objective:
Medical providers may order CD4 tests for people with HIV (PWH) without providing HIV care. We hypothesized that including CD4 tests as a proxy for HIV medical visits misclassifies PWH on linkage to, receipt of, and retention in care.
Methods:
We used 2021 HIV surveillance data from New York City for analysis and compared 2 proxies for HIV medical visits—one included the presence of either CD4 or viral load tests and the other included viral load tests—to measure linkage to care among people newly diagnosed with HIV and receipt of and retention in care among PWH.
Results:
In New York City in 2021, 1146 of 1399 people newly diagnosed with HIV (81.9%) had timely linkage to care when either CD4 or viral load tests were used as proxies; 1118 of 1399 (79.9%) people newly diagnosed with HIV had timely linkage to care when only viral load tests were used. A total of 69 982 of 76 338 PWH (91.7%) were classified as having received HIV care in New York City in 2021 when either CD4 or viral load tests were used; 69 088 of 76 388 PWH (90.5%) were classified as having received HIV care when only viral load tests were used.
Conclusions:
Including CD4 tests as proxies for HIV medical visits misclassifies linkage to care among people newly diagnosed with HIV, and receipt of and retention in care among PWH. We recommend using viral load tests only as proxies to monitor progress toward goals of the National HIV/AIDS Strategy.
Keywords: HIV, CD4, viral load, surveillance, access to care
People with HIV (PWH) who receive timely linkage to care after HIV diagnosis and ongoing, regularly scheduled care have significantly lower HIV viral loads, higher CD4 cell counts, and reduced morbidity and mortality when compared with those who are not promptly linked to care or not retained in HIV care.1–4
Routine HIV medical care refers to services received during a clinical encounter with a primary HIV medical provider.5,6 Because of the difficulties in obtaining data on HIV medical visits, proxy measures from HIV surveillance data have been used by public health agencies to monitor linkage to care among people newly diagnosed with HIV and retention in care among PWH.7–12 CD4 (cell count or percentage) and viral load tests are 2 common proxies used for HIV medical visits in the United States and around the globe.13–19 Using either CD4 or viral load tests as proxies for routine HIV medical visits, the Centers for Disease Control and Prevention (CDC) reported that in the United States in 2021, 81.9% of people newly diagnosed with HIV were linked to HIV care within 1 month of diagnosis, 75.3% of PWH received HIV care, and 53.9% of PWH were retained in HIV care.20 The latest national guidelines for the clinical care and treatment of PWH have placed less emphasis on CD4 testing than previous guidelines because it is no longer required to guide antiretroviral treatment (ART) eligibility and provides limited information on patients who are consistently virally suppressed and have already experienced ART-related immune reconstitution (ie, CD4 count >500 cells/mm3).21
All HIV viral load tests are HIV-related but not all CD4 tests are. A CD4 cell count measure can be used to understand the progression of HIV infection but could also be used for non–HIV-related clinical management for PWH (eg, monitoring treatment after an organ transplant, helping diagnose various types of lymphoma). Some non-HIV medical providers may order CD4 tests to check the immune system of PWH without providing HIV medical care.22 There is no way to tease out CD4 tests that are part of HIV medical care visits versus those that are not part of HIV medical care visits in HIV registry. We hypothesized that including CD4 tests as proxies for HIV medical visits would contribute to misclassification of PWH on linkage to care, receipt of care, and retention in care and inflate population estimates.
Methods
Data Source
The data source for this analysis was the New York City HIV Registry (hereinafter, the Registry). AIDS diagnoses have been reportable in New York State since 1981, HIV diagnoses since 2000, and HIV-related tests—including CD4 counts and percentages, viral loads, and nucleotide sequences—since 2005. HIV diagnoses in New York City are reported directly to the New York City Department of Health and Mental Hygiene (NYC Health Department) and entered into the Registry, and HIV-related laboratory tests, including all CD4 and HIV viral load tests, on New York City residents are reported electronically to the New York State Department of Health via the Electronic Clinical Laboratory Reporting System and then sent to the NYC Health Department to be included in the Registry.
The Registry is continuously updated with new, deduplicated diagnoses and laboratory results. All incoming health care provider and laboratory reports that do not match an existing registry record initiate an investigation with medical record review to ascertain whether the case is confirmed as a new or previous diagnosis. Data collected for surveillance and partner notification include the patient’s name, date of birth, sex assigned at birth, gender, race or ethnicity, transmission category, residential address, and HIV-related clinical information. A patient’s vital status is verified by quarterly matches with the New York City death registry and yearly matches with the National Death Index and the Social Security Death Master File. As of December 31, 2021, the Registry contained a cumulative total of more than 270 000 cases (alive and dead) and more than 13 million laboratory test results.
The analysis used surveillance data and was a public health practice activity, not human subjects research. Thus, it was exempt from institutional review board approval.
Outcome Measures
When allowing either CD4 or viral load tests as proxy measures for HIV medical visits, we used CDC’s definitions: (1) timely linkage to care as ≥1 CD4 (count or percentage) or viral load test performed ≤1 month after HIV diagnosis, including tests performed on the same date as the date of diagnosis; (2) receipt of HIV care as ≥1 CD4 or viral load test performed in a calendar year; and (3) retention in care as ≥2 CD4 or viral load tests performed ≥3 months apart in a calendar year.20
When including only viral load tests as proxies for HIV medical visits, we used the following definitions: (1) timely linkage to care as ≥1 viral load test performed ≤1 month after HIV diagnosis, including tests performed on the same date as the date of diagnosis; (2) receipt of HIV care as ≥1 viral load test performed in a calendar year; and (3) retention in care as ≥2 viral load tests performed ≥3 months apart in a calendar year.
Analysis Populations
For the linkage-to-care analysis, we included people who were newly diagnosed with HIV in New York City in 2021, aged ≥18 years, and New York City residents at the time of diagnosis. For the receipt-of-care and retention-in-care analyses, we included PWH who were diagnosed with HIV prior to 2021, aged ≥18 years, New York City residents, and alive at the end of 2021. The number of PWH was estimated using a previously described statistical weighting method on the HIV registry data.12,23
This method treated the New York City HIV laboratory data reporting system as a special annual population-based survey, lasting from January 1 to December 31 each year. Patients who had at least 1 CD4 or viral load test in the year were considered participants in the annual survey. Each participant was then given a weight equal to the inverse of the probability that the patient had a CD4 or viral load test in New York City in the year. The probability was calculated based on the time interval between the last care visit before January 1 of the year or date of diagnosis if there were no care visits before the year and the first care visit was in the year. If a patient’s interval was less than or equal to 1 year, meaning that the patient was in regular care and definitely included in the annual survey in the year with a probability of 100%, the patient received a weight of 1. If a patient’s interval was greater than 1 year, the patient received a weight equal to the time interval in years. For example, if a patient had his last care visit 2 and half years before his first care visit in the year, he received a weight of 2.5, and the patient represented not only himself, but also 1 and a half out-of-care patients.
The populations are stratified by gender (men, women, and transgender), race or ethnicity (Black, Latino/Hispanic, White, Asian/Pacific Islander, American Indian/Alaska Native, and Multiracial), age (18–24, 25–34, 35–44, 45–54, and ≥65 years), and transmission category (men who have sex with men [MSM], people who inject drugs [PWID], MSM/PWID, Heterosexual, and Other/unknown). The population for receipt of and retention in care analyses is also stratified by year of diagnosis (pre-1991, 1991–1995, 1996–2000, 2001–2005, 2006–2010, 2011–2015, and 2016–2021).
Statistical Analysis
We reported the number and proportion for timely linkage to care among people who were newly diagnosed with HIV in New York City in 2021, and the numbers and proportions for receipt of care and retention in care among PWH residing in New York City in 2021, overall and by gender, race or ethnicity, age, and transmission category. We compared the numbers and proportions between the 2 definitions: definition #1 using either CD4 or viral load tests as proxies for HIV medical visits and definition #2 using viral load tests only as proxies for HIV medical visits.
Results
A total of 1399 people were newly diagnosed with HIV in New York City in 2021 (Table 1). The number (percentage) of people with timely linkage to care was 1146 (81.9%) when either CD4 or viral load tests were used as proxies for HIV medical visits (definition #1) and 1118 (79.9%), 28 people fewer and 2.0 percentage points lower, when only viral load tests were used as proxies (definition #2). The differences by proxy definition were relatively small in all gender, race or ethnicity, age, and transmission groups, ranging from no differences (transgender, American Indian/Alaska Native, multiracial, aged ≥65 years, PWID, and MSM/PWID groups) to 4.9 percentage points among people aged 55 to 64 years.
Table 1.
Linkage to care among people newly diagnosed with HIV, by HIV medical visit proxy definition, New York City, 2021
| Characteristic | Total, no. | Definition #1 (≥1 CD4/viral load test within 1 month of diagnosis), no. (%) | Definition #2 (≥1 viral load test within 1 month of diagnosis), no. (%) | Misclassification, no. | Overestimation, percentage point |
|---|---|---|---|---|---|
| Total | 1399 | 1146 (81.9) | 1118 (79.9) | 28 | 2.0 |
| Gender | |||||
| Men | 1107 | 907 (81.9) | 883 (79.8) | 24 | 2.2 |
| Women | 243 | 199 (81.9) | 195 (80.2) | 4 | 1.6 |
| Transgender | 49 | 40 (81.6) | 40 (81.6) | 0 | 0 |
| Race and ethnicity | |||||
| Black | 619 | 493 (79.6) | 481 (77.7) | 12 | 1.9 |
| Latino/Hispanic | 523 | 438 (83.7) | 427 (81.6) | 11 | 2.1 |
| White | 159 | 135 (84.9) | 132 (83.0) | 3 | 1.9 |
| Asian/Pacific Islander | 81 | 69 (85.2) | 67 (82.7) | 2 | 2.5 |
| American Indian/Alaska Native | 1 | 0 | 0 | 0 | 0 |
| Multiracial | 16 | 11 (68.8) | 11 (68.8) | 0 | 0 |
| Age, y | |||||
| 18–24 | 252 | 212 (84.1) | 205 (81.3) | 7 | 2.8 |
| 25–34 | 537 | 441 (82.1) | 434 (80.8) | 7 | 1.3 |
| 35–44 | 285 | 230 (80.7) | 224 (78.6) | 6 | 2.1 |
| 45–54 | 166 | 137 (82.5) | 135 (81.3) | 2 | 1.2 |
| 55–64 | 123 | 100 (81.3) | 94 (76.4) | 6 | 4.9 |
| ≥65 | 36 | 26 (72.2) | 26 (72.2) | 0 | 0 |
| Transmission category | |||||
| MSM | 697 | 609 (87.4) | 596 (85.5) | 13 | 1.9 |
| PWID | 10 | 7 (70.0) | 7 (70.0) | 0 | 0 |
| MSM/PWID | 21 | 16 (76.2) | 16 (76.2) | 0 | 0 |
| Heterosexual | 213 | 181 (85.0) | 177 (83.1) | 4 | 1.9 |
| Other/unknown | 458 | 333 (72.7) | 322 (70.3) | 11 | 2.4 |
Abbreviations: MSM, men who have sex with men; PWID, people who inject drugs.
A total of 76 338 PWH were living in New York City at the end of 2021 (Table 2). The number (percentage) of PWH classified as having received HIV care in New York City in 2021 was 69 982 (91.7%) when either CD4 or viral load tests were used as proxies for HIV medical visits (definition #1) and 69 088 (90.5%), 894 people fewer and 1.2 percentage points lower, when only viral load tests were used as proxies (definition #2). The differences by proxy definition were relatively small by gender, race or ethnicity, age, transmission, and year-of-diagnosis groups, varying from 0.7 percentage points among transgender people to 2.6 percentage points among American Indian/Alaska Native people. The misclassification is similar by year of diagnosis between 0.8 percentage points among people diagnosed prior to 1991 and 1.3 percentage points among people diagnosed during 2011–2015.
Table 2.
Receipt of care among people with HIV, by HIV medical visit proxy definition, New York City, 2021
| Characteristic | Total, no.a | Definition #1 (≥1 CD4/viral load test in 2021), no. (%) | Definition #2 (≥1 viral load test in 2021), no. (%) | Misclassification, no. | Overestimation, percentage point |
|---|---|---|---|---|---|
| Total | 76 338 | 69 982 (91.7) | 69 088 (90.5) | 894 | 1.2 |
| Gender | |||||
| Men | 53 594 | 49 041 (91.5) | 48 416 (90.3) | 625 | 1.2 |
| Women | 20 958 | 19 292 (92.1) | 19 035 (90.8) | 257 | 1.2 |
| Transgender | 1786 | 1649 (92.3) | 1637 (91.7) | 12 | 0.7 |
| Race and ethnicity | |||||
| Black | 34 998 | 31 704 (90.6) | 31 276 (89.4) | 428 | 1.2 |
| Latino/Hispanic | 26 733 | 24 804 (92.8) | 24 494 (91.6) | 310 | 1.2 |
| White | 11 924 | 11 024 (92.5) | 10 898 (91.4) | 126 | 1.1 |
| Asian/Pacific Islander | 2085 | 1907 (91.5) | 1885 (90.4) | ||
| American Indian/Alaska Native | 193 | 170 (88.2) | 165 (85.6) | ||
| Multiracial | 324 | 297 (91.6) | 294 (90.7) | 3 | 0.9 |
| Unknown | 81 | 76 (93.5) | 76 (93.5) | 0 | 0.0 |
| Age, y | |||||
| 18–24 | 1062 | 980 (92.3) | 963 (90.7) | 17 | 1.6 |
| 25–34 | 10 356 | 9315 (90.0) | 9178 (88.6) | 137 | 1.3 |
| 35–44 | 13 621 | 12 195 (89.5) | 12 027 (88.3) | 168 | 1.2 |
| 45–54 | 16 368 | 15 091 (92.2) | 14 925 (91.2) | 166 | 1.0 |
| 55–64 | 22 661 | 20 962 (92.5) | 20 737 (91.5) | 225 | 1.0 |
| ≥65 | 12 271 | 11 439 (93.2) | 11 258 (91.7) | 181 | 1.5 |
| Transmission category | |||||
| MSM | 34 029 | 31 447 (92.4) | 31 098 (91.4) | 349 | 1.0 |
| PWID | 6750 | 6295 (93.3) | 6199 (91.8) | 96 | 1.4 |
| MSM/PWID | 2173 | 1966 (90.5) | 1936 (89.1) | 30 | 1.4 |
| Heterosexual | 16 789 | 15 398 (91.7) | 15 191 (90.5) | 207 | 1.2 |
| Perinatal | 1460 | 1261 (86.3) | 1242 (85.0) | 19 | 1.3 |
| Other/unknown | 15 137 | 13 615 (89.9) | 13 422 (88.7) | 193 | 1.3 |
| Year of diagnosis | |||||
| Pre-1991 | 5061 | 4634 (91.6) | 4593 (90.8) | 41 | 0.8 |
| 1991–1995 | 8790 | 8263 (94.0) | 8171 (93.0) | 92 | 1.0 |
| 1996–2000 | 14 834 | 13 781 (92.9) | 13 598 (91.7) | 183 | 1.2 |
| 2001–2005 | 13 891 | 12 688 (91.3) | 12 524 (90.2) | 164 | 1.2 |
| 2006–2010 | 12 891 | 11 588 (89.9) | 11 429 (88.7) | 159 | 1.2 |
| 2011–2015 | 11 798 | 10 597 (89.8) | 10 446 (88.5) | 151 | 1.3 |
| 2016–2021 | 9073 | 8431 (92.9) | 8327 (91.8) | 104 | 1.1 |
Abbreviations: MSM, men who have sex with men; PWID, people who inject drugs.
The number of people with HIV is estimated and not a whole number, but weighted. Because of rounding of weights, the sum may not equal to the total and the proportion may not equal to the value from simple calculations based on the numerator and denominator in the table.
The number (percentage) of PWH retained in HIV care in New York City in 2021 was 54 629 (71.6%) when either CD4 or viral load tests were used as proxies for HIV medical visits (definition #1) and 52 662 (69.0%), 1967 people fewer and 2.6 percentage points lower, when only viral load tests were used as proxies (definition #2) (Table 3). The differences by proxy definition were larger than those for receipt of care but still relatively small by gender, race or ethnicity, age, transmission category, and year-of-diagnosis groups, varying from 1.6 percentage points among transgender people to 3.6 percentage points among American Indian/Alaska Native people, adults aged 18 to 24 years, and PWID. Unlike receipt of care, there is a pattern in the misclassification of retention in care with less misclassification in more recent years. The overestimation decreased from 3.4 percentage points among people diagnosed prior to 1991 to 2.3 percentage points among people diagnosed between 2016–2021.
Table 3.
Retention in care among people with HIV, by HIV care proxy definition, New York City, 2021
| Characteristic | Total, no.a | Definition #1 (≥2 CD4/viral load tests ≥3 months apart in 2021), no. (%) | Definition #2 (≥2 viral load tests ≥3 months apart in 2021), no. (%) | Misclassification. no. | Overestimation, percentage point |
|---|---|---|---|---|---|
| Total | 76 338 | 54 629 (71.6) | 52 662 (69.0) | 1967 | 2.6 |
| Gender | |||||
| Men | 53 594 | 37 952 (70.8) | 36 596 (68.3) | 1356 | 2.5 |
| Women | 20 958 | 15 352 (73.3) | 14 769 (70.5) | 583 | 2.8 |
| Transgender | 1786 | 1325 (74.2) | 1297 (72.6) | 28 | 1.6 |
| Race and ethnicity | |||||
| Black | 34 998 | 24 660 (70.5) | 23 766 (67.9) | 894 | 2.6 |
| Latino/Hispanic | 26 733 | 19 892 (74.4) | 19 153 (71.6) | 739 | 2.8 |
| White | 11 924 | 8184 (68.6) | 7906 (66.3) | 278 | 2.3 |
| Asian/Pacific Islander | 2085 | 1500 (71.9) | 1461 (70.1) | 39 | 1.9 |
| American Indian/Alaska Native | 193 | 116 (60.2) | 109 (56.5) | ||
| Multiracial | 324 | 215 (66.3) | 205 (63.3) | 10 | 3.1 |
| Unknown | 81 | 62 (76.3) | 62 (76.3) | 0 | 0.0 |
| Age, y | |||||
| 18–24 | 1062 | 729 (68.7) | 691 (65.1) | 38 | 3.6 |
| 25–34 | 10 356 | 6720 (64.9) | 6466 (62.4) | 254 | 2.5 |
| 35–44 | 13 621 | 8949 (65.7) | 8634 (63.4) | 315 | 2.3 |
| 45–54 | 16 368 | 11 713 (71.6) | 11 351 (69.4) | 362 | 2.2 |
| 55–64 | 22 661 | 17 100 (75.5) | 16 522 (72.9) | 578 | 2.6 |
| ≥65 | 12 271 | 9418 (76.7) | 8998 (73.3) | 420 | 3.4 |
| Transmission category | |||||
| MSM | 34 029 | 24 054 (70.7) | 23 304 (68.5) | 750 | 2.2 |
| PWID | 6750 | 5225 (77.4) | 4980 (73.8) | 245 | 3.6 |
| MSM/PWID | 2173 | 1565 (72.0) | 1504 (69.2) | 61 | 2.8 |
| Heterosexual | 16 789 | 12 160 (72.4) | 11 684 (69.6) | 476 | 2.8 |
| Perinatal | 1460 | 942 (64.5) | 896 (61.3) | 46 | 3.1 |
| Other/unknown | 15 137 | 10 683 (70.6) | 10 294 (68.0) | 389 | 2.6 |
| Year of diagnosis | |||||
| Pre-1991 | 5061 | 3820 (75.5) | 3648 (72.1) | 172 | 3.4 |
| 1991–1995 | 8790 | 6841 (77.8) | 6605 (75.1) | 236 | 2.7 |
| 1996–2000 | 14 834 | 11 045 (74.5) | 10 643 (71.7) | 402 | 2.7 |
| 2001–2005 | 13 891 | 9955 (71.7) | 9595 (69.1) | 360 | 2.6 |
| 2006–2010 | 12 891 | 8747 (67.9) | 8439 (65.5) | 308 | 2.4 |
| 2011–2015 | 11 798 | 7815 (66.2) | 7535 (63.9) | 280 | 2.4 |
| 2016–2021 | 9073 | 6406 (70.6) | 6197 (68.3) | 209 | 2.3 |
Abbreviations: MSM, men who have sex with men; PWID, people who inject drugs.
The number of people with HIV is estimated and not a whole number, but weighted. Because of rounding of weights, the sum may not equal to the total and the proportion may not equal to the value from simple calculations based on the numerator and denominator in the table.
Discussion
Our analysis showed that in New York City, when either CD4 or viral load tests were used as proxies for HIV medical visits, 28 people were misclassified as linked to care within 1 month of diagnosis, 894 people were misclassified as having received HIV care in 2021, 1967 people were misclassified as retained in HIV care in 2021, and the estimated proportions for linkage to care, receipt of care, and retention in care were 2.0, 1.2, and 2.6 percentage points higher, respectively, than the estimated proportions when using only viral load tests as proxies.
The standard definition (definition #1) for linkage to care included 28 people who each received exactly 1 CD4 test within 1 month of diagnosis. Among them, only 1 patient’s CD4 test was ordered by an infectious disease physician; the others included 5 by a family medicine physician, 5 by a nurse practitioner, 4 by an internal medicine physician, 3 by an emergency medicine physician, 1 by an addiction medicine physician, 1 by an adolescent medicine physician, 1 by an anesthesiologist, 1 by a cardiologist, 1 by a gastroenterologist, 1 by a gynecologist, 1 by a neurologist, 1 by a preventive medicine physician, 1 by a psychiatrist, and 1 by a physician assistant. The first viral load tests for these patients were ordered by an infectious disease physician more than 1 month after HIV diagnosis, which likely corresponded to their actual start of HIV care.
Among those classified under the standard definition as having received HIV care in 2021, a total of 894 PWH had at least 1 CD4 test and no viral load tests for the entire duration of 2021; 623 of them had only 1 CD4 test and the other 271 had more than 1 CD4 test. Because it is unlikely a person in New York City can be receiving HIV care without a single viral load test being ordered by their primary medical provider in an entire calendar year, given current recommendations for laboratory testing schedules, we regard these patients as having been misclassified under definition #1.
Among those classified under the standard definition as having been retained in care in 2021, 183 PWH had CD4 tests only and 1784 had only one viral load test for the entire duration of 2021. We observed a pattern of less misclassification among people recently diagnosed that the overestimation decreased from 3.4 percentage points among people diagnosed prior to 1991 to 2.3 percentage points among people diagnosed between 2016–2021. The pattern is likely caused by more frequent viral load monitoring among people recently diagnosed with HIV than people diagnosed earlier.21 People diagnosed earlier are more likely to have consistently suppressed viral load and have only one viral load test in a calendar year (and one or more CD4 tests at different times in the year). Therefore, they are more likely to be classified as retained in care under the standard definition using either CD4 or viral load tests as proxies for HIV medical visits, but not retained in care under the new definition using viral load tests only as proxies for HIV medical visits. For example, a patient who had a CD4 test in March and a viral load test in September would be classified as retained in care under the standard definition, but not retained in care under the new definition. This raises a question whether the measure of retention in care, which requires two viral load tests at least 3 months apart in a year, is too strict, given the fact that more and more PWH are able to maintain their viral suppression and require less frequent viral load monitoring. We recommend using receipt of HIV care (≥1 viral load test performed in a calendar year), not retention in care (≥2 viral load tests performed ≥3 months apart in a calendar year) to monitor HIV care among PWH.24
Limitations
Our analysis had several limitations. First, without data on HIV medical visits (ie, the gold standard), we were comparing 2 approaches both using proxies for medical visits, not knowing which approach is more accurate. Both false positives and false negatives affect accuracy, which is measured by the formula, accuracy = (true positives + true negatives)/(true positives + true negatives + false positives + false negatives). However, our analysis was focused more on false positives and we were able to demonstrate that using viral load tests only as proxies produces fewer false positives than using either CD4 or viral load tests as proxies. Second, we were unable to extract data from medical records to confirm whether the 28 people newly diagnosed with HIV in 2021 without a viral load test within 1 month of diagnosis and 894 PWH without a viral load test in 2021 had viral load tests during the analysis period that were not reported to the NYC Department of Health. We were able to obtain viral load tests after 1 month of diagnosis among these 28 new diagnoses and viral load tests from other patients who received care at the same facilities as the 894 PWH without a viral load test in 2021. Therefore, we believe the absence of viral load tests among these 28 and 894 patients reflected a lack of viral load monitoring for the periods and patients in question rather than the result of incomplete reporting. Third, we did not include HIV genotype tests as proxies for HIV medical visits for the following reasons: (1) HIV genotype tests are not reportable in all jurisdictions in the United States and our recommendation is for a standard definition that can be consistently applied; (2) it is rare for PWH to have an HIV genotype test without an accompanying viral load test result reported to the Health Department; and (3) a small number of such “false negatives” misclassified as out of care (33 of 69 121 PWH in New York City in 2021) would have little effect on the measure of retention in care at the population level. Furthermore, there is little harm and may be more benefit if such people receive outreach by an HIV care re-engagement team from a public health agency. Fourth, measures relying solely on viral load testing may not adequately reflect linkage to care shortly after HIV diagnosis. Although CD4 testing is no longer required to determine eligibility for ART, because ART is recommended for all PWH, CD4 counts remain crucial for evaluating immune status and risk stratification for opportunistic infections, especially in the early stage after HIV diagnosis.21,25 Thus, although our recommendation to require viral load measurements is more rigorous than the current practice of requiring either CD4 or viral load measurements, it may not be aligned with clinical care standards requiring both CD4 and viral load measurements. For our analysis, if both CD4 and viral load measurements are required, the proportion of people linked to care within 1 month of diagnosis is further reduced from 79.9% to 74.5%. However, CD4 testing among people who have been previously diagnosed, are consistently virally suppressed, and have already experienced ART-related immune reconstitution (ie, CD4 count >500 cells/mm3) is not as important and might not be as widely practiced as CD4 testing among people newly diagnosed with HIV. Requiring both CD4 and viral load measurements would underestimate receipt of and retention in care among PWH. To be consistent across 3 HIV care measures and 2 populations—linkage to care among people newly diagnosed with HIV and receipt of and retention in care among PWH—we recommend using viral load measurements only as proxies for medical visits.
Conclusion
Not all CD4 tests are HIV-related and non-HIV medical providers may order CD4 tests on PWH without providing HIV care. Including either CD4 or viral load tests as proxies for HIV medical visits risks misclassifying the status of some PWH on linkage to care, receipt of care, and retention in care, and results in some inflation of population estimates. Although the magnitude of overestimation on a population level is relatively small (2.0 percentage points for linkage to care, 1.2 percentage points for receipt of care, and 2.6 percentage points for retention in care), the numbers are not: 28 people newly diagnosed with HIV were misclassified as linked to care within 1 month of diagnosis, 894 PWH were misclassified as having received HIV care in 2021, and 1967 PWH were misclassified as retained in HIV care in 2021. Misclassifications can have serious consequences at the individual level. If surveillance data are used to identify out-of-care individuals to contact for re-engagement in care, PWH misclassified as being in care would be excluded from such re-engagement efforts. The misclassification was also found to be more common in populations with limited access to health care (eg, American Indian/Alaska Native people, PWID). Missing more people from these populations in our linkage and re-engagement efforts could exacerbate existing HIV outcome disparities. Therefore, we recommend that surveillance programs use viral load tests only as proxies for HIV medical care to identify people in need of linkage and relinkage to care and to monitor progress toward National HIV/AIDS Strategy goals.26 Of note, our recommendation applies only to public health monitoring and evaluation and does not apply to clinical management. Despite the shift to universal ART (regardless of CD4 count), the clinical guidelines still support both CD4 and viral load testing because of their clinical implications for overall care management, especially early after HIV diagnosis.21
Acknowledgments
The authors thank Sarah Braunstein, PHD, MPH, Celia Quinn, MD, MPH, Matthew Kuehnert, MD, and Hannah Helmy, PhD (all with the New York City Department of Health and Mental Hygiene), for their review and comments on this article.
Funding
The authors received the following financial support for the research, authorship, and/or publication of this article: This work was supported in part by the Centers for Disease Control and Prevention (NU62PS924575 and NU62PS924626). Denis Nash was supported by the Einstein, Rockefeller, CUNY Center for AIDS Research, National Institutes of Health grant no. P30 AI124414.
Footnotes
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.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
- 1.Lundgren JD, Babiker AG, Gordin F, et al. Initiation of antiretroviral therapy in early asymptomatic HIV infection. N Engl J Med. 2015;373(9):795–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mugavero MJ, Amico KR, Westfall AO, et al. Early retention in HIV care and viral load suppression: implications for a test and treat approach to HIV prevention. J Acquir Immune Defic Syndr. 2012;59(1):86–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Robertson M, Laraque F, Mavronicolas H, Braunstein S, Torian L. Linkage and retention in care and the time to HIV viral suppression and viral rebound - New York City. AIDS Care. 2015;27(2):260–267. [DOI] [PubMed] [Google Scholar]
- 4.Ulloa AC, Puskas C, Yip B, et al. Retention in care and mortality trends among patients receiving comprehensive care for HIV infection: a retrospective cohort study. CMAJ Open. 2019;7(2):E236–e245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mugavero MJ, Westfall AO, Zinski A, et al. Measuring retention in HIV care: the elusive gold standard. J Acquir Immune Defic Syndr. 2012;61(5):574–580. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Brazier E, Tymejczyk O, Wools-Kaloustian K, et al. Long-term HIV care outcomes under universal HIV treatment guidelines: A retrospective cohort study in 25 countries. PLoS Med. 2024;21(3):e1004367. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dean BB, Hart RL, Buchacz K, Bozzette SA, Wood K, Brooks JT. HIV laboratory monitoring reliably identifies persons engaged in care. J Acquir Immune Defic Syndr. 2015;68(2):133–139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rebeiro PF, Althoff KN, Lau B, et al. Laboratory measures as proxies for primary care encounters: implications for quantifying clinical retention among HIV-infected adults in North America. Am J Epidemiol. 2015;182(11):952–960. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lubelchek RJ, Finnegan KJ, Hotton AL, et al. Assessing the use of HIV surveillance data to help gauge patient retention-in-care. J Acquir Immune Defic Syndr. 2015;69 Suppl 1(0 1):S25–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Pati R, Robbins RS, Braunstein SL. Validation of retention in HIV care status using the New York City HIV surveillance registry and clinical care data from a large HIV care center. J Public Health Manag Pract. 2017;23(6):564–570. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Xia Q, Zhong Y, Wiewel EW, Braunstein SL, Torian LV. Linkage to care after HIV diagnosis in New York City: better than we thought. J Acquir Immune Defic Syndr. 2017;76(1):e18–e21. [DOI] [PubMed] [Google Scholar]
- 12.Xia Q, Kersanske LS, Wiewel EW, Braunstein SL, Shepard CW, Torian LV. Proportions of patients with HIV retained in care and virally suppressed in New York City and the United States: higher than we thought. J Acquir Immune Defic Syndr. 2015;68(3):351–358. [DOI] [PubMed] [Google Scholar]
- 13.Torian LV, Wiewel EW, Liu KL, Sackoff JE, Frieden TR. Risk factors for delayed initiation of medical care after diagnosis of human immunodeficiency virus. Arch Intern Med. 2008;168(11):1181–1187. [DOI] [PubMed] [Google Scholar]
- 14.Dombrowski JC, Buskin SE, Bennett A, Thiede H, Golden MR. Use of multiple data sources and individual case investigation to refine surveillance-based estimates of the HIV care continuum. J Acquir Immune Defic Syndr. 2014;67(3):323–330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gray KM, Cohen SM, Hu X, Li J, Mermin J, Hall HI. Jurisdiction level differences in HIV diagnosis, retention in care, and viral suppression in the United States. J Acquir Immune Defic Syndr. 2014;65(2):129–132. [DOI] [PubMed] [Google Scholar]
- 16.Nosyk B, Montaner JSG, Colley G, et al. The cascade of HIV care in British Columbia, Canada, 1996–2011: a population-based retrospective cohort study. Lancet Infect Dis. 2014;14(1):40–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Irvine MK, Chamberlin SA, Robbins RS, et al. Improvements in HIV care engagement and viral load suppression following enrollment in a comprehensive HIV care coordination program. Clin Infect Dis. 2015;60(2):298–310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Chkhartishvili N, Sharavdze L, Chokoshvili O, DeHovitz JA, del Rio C, Tsertsvadze T. The cascade of care in the Eastern European country of Georgia. HIV Med. 2015;16(1):62–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rice BD, Yin Z, Brown AE, et al. Monitoring of the HIV epidemic using routinely collected data: the case of the United Kingdom. AIDS and Behavior. 2017;21(1):83–90. [DOI] [PubMed] [Google Scholar]
- 20.Centers for Disease Control and Prevention. Monitoring selected national HIV prevention and care objectives by using HIV surveillance data United States and 6 U.S. dependent areas, 2021. HIV Surveillance Supplemental Report. 2023;28(4). [Google Scholar]
- 21.Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in adults and adolescents with HIV. Department of Health and Human Services;2024. [Google Scholar]
- 22.Sabharwal CJ, Braunstein SL, Robbins RS, Shepard CW. Optimizing the use of surveillance data for monitoring the care status of persons recently diagnosed with HIV in NYC. J Acquir Immune Defic Syndr. 2014;65(5):571–578. [DOI] [PubMed] [Google Scholar]
- 23.Xia Q, Neaigus A, Bernard MA, Raj-Singh S, Shepard CW. Constructing a representative sample of out-of-care HIV patients from a representative sample of in-care patients. Int J STD AIDS. 2016;27(12):1086–1092. [DOI] [PubMed] [Google Scholar]
- 24.Xia Q, Wiewel EW, Torian LV. Comparison of single-visit and multiple-visit measures of retention in care for HIV monitoring and evaluation. J Acquir Immune Defic Syndr. 2016;71(2):e59–62. [DOI] [PubMed] [Google Scholar]
- 25.Brazier E, Tymejczyk O, Zaniewski E, et al. Effects of national adoption of treat-all guidelines on pre-antiretroviral therapy (ART) CD4 testing and viral load monitoring after ART initiation: a regression discontinuity analysis. Clinical Infectious Diseases. 2021;73(6):e1273–e1281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.U.S. Department of Health and Human Services. HIV National Strategic Plan for the United States: A Roadmap to End the Epidemic 2021–2025. Washington, DC.2021. [Google Scholar]