Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=caic20

Download by: [UCSF Library] Date: 04 May 2016, At: 23:52

AIDS CarePsychological and Socio-medical Aspects of AIDS/HIV

ISSN: 0954-0121 (Print) 1360-0451 (Online) Journal homepage: http://www.tandfonline.com/loi/caic20

Movement between facilities for HIV care amonga mobile population in Kenya: transfer, loss tofollow-up, and reengagement

Matthew D. Hickey, Dan Omollo, Charles R. Salmen, Brian Mattah, CinthiaBlat, Gor Benard Ouma, Kathryn J. Fiorella, Betty Njoroge, Monica Gandhi,Elizabeth A. Bukusi, Craig R. Cohen & Elvin H. Geng

To cite this article: Matthew D. Hickey, Dan Omollo, Charles R. Salmen, Brian Mattah, CinthiaBlat, Gor Benard Ouma, Kathryn J. Fiorella, Betty Njoroge, Monica Gandhi, Elizabeth A. Bukusi,Craig R. Cohen & Elvin H. Geng (2016): Movement between facilities for HIV care among amobile population in Kenya: transfer, loss to follow-up, and reengagement, AIDS Care, DOI:10.1080/09540121.2016.1179253

To link to this article: http://dx.doi.org/10.1080/09540121.2016.1179253

Published online: 04 May 2016.

Submit your article to this journal

View related articles

View Crossmark data

Movement between facilities for HIV care among a mobile population in Kenya:transfer, loss to follow-up, and reengagementMatthew D. Hickeya,b,c, Dan Omollob, Charles R. Salmenb,c,d, Brian Mattahb, Cinthia Blate, Gor Benard Oumab,Kathryn J. Fiorellab,f, Betty Njorogeg, Monica Gandhih, Elizabeth A. Bukusig, Craig R. Cohene,i and Elvin H. Gengh

aDivision of General Internal Medicine, University of California, San Francisco (UCSF), San Francisco, CA, USA; bMfangano Island Research Group,Organic Health Response, Homa Bay County, Kenya; cMicroclinic International (MCI), San Francisco, CA, USA; dDepartment of Family andCommunity Medicine, University of Minnesota, Minneapolis, MN, USA; eGlobal Health Sciences, University of California, San Francisco (UCSF),San Francisco, CA, USA; fAtkinson Center for a Sustainable Future, Cornell University, Ithaca, NY, USA; gCentre for Microbial Research,Kenya Medical Research Institute, Nairobi, Kenya; hHIV-ID-Global Medicine Division, University of California, San Francisco (UCSF), San Francisco,CA, USA; iDepartment of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco (UCSF), San Francisco,CA, USA

ABSTRACTHIV treatment is life-long, yet many patients travel or migrate for their livelihoods, risking treatmentinterruption. We examine timely reengagement in care among patients who transferred-out orwere lost-to-follow-up (LTFU) from a rural HIV facility. We conducted a cohort study among 369adult patients on antiretroviral therapy between November 2011 and November 2013 onMfangano Island, Kenya. Patients who transferred or were LTFU (i.e., missed a scheduledappointment by ≥90 days) were traced to determine if they reengaged or accessed care atanother clinic. We report cumulative incidence and time to reengagement using Coxproportional hazards models adjusted for patient demographic and clinical characteristics.Among 369 patients at the clinic, 23(6%) requested an official transfer and 78(21%) were LTFU.Among official transfers, cumulative incidence of linkage to their destination facility was 91% atthree months (95%CI (confidence intervals) 69–98%). Among LTFU, cumulative incidence ofreengagement in care at the original or a new clinic was 14% at three months (95%CI 7–23%)and 60% at six months (95%CI 48–69%). In the adjusted Cox model, patients who left with anofficial transfer reengaged in care six times faster than those who did not (adjusted hazard ratio6.2, 95%CI 3.4–11.0). Patients who left an island-based HIV clinic in Kenya with an official transferletter reengaged in care faster than those who were LTFU, although many in both groups hadtreatment gaps long enough to risk viral rebound. Better coordination of transfers betweenclinics, such as assisting patients with navigating the process or improving inter-cliniccommunication surrounding transfers, may reduce delays in treatment during transfer andimprove overall clinical outcomes.

ARTICLE HISTORYReceived 4 November 2015Accepted 12 April 2016

KEYWORDSHIV; transfer; lost to follow-up; stigma; resource-limitedsetting

Introduction

Individuals receiving care and treatment for HIV inresource-limited settings face a unique challenge: whilethey must engage in care over long periods of time tooptimize health outcomes, they often require a highdegree of mobility to meet livelihood demands (Camlin,Snow, & Hosegood, 2014; Marson et al., 2013; Wareet al., 2013). Furthermore, the increasing numbers of“decentralized” clinics opening in rural areas presentpatients with new choices in care and incentives to trans-fer, such as shorter transit times, shorter waiting times,or friendlier staff (Geng et al., 2010; Ware et al., 2013).Thus, the reality of treatment for HIV-infected patientsoften includes accessing multiple sites for clinical ser-vices over time (Nglazi et al., 2011).

Patient movement between sites can occur in ways thatare either known (i.e., through official transfer request) orunknown (i.e., patients are “lost to follow-up” or LTFU) tothe original clinic. There is considerable interest in under-standing outcomes among patients who cease accessingcare through either of these mechanisms (Brinkhof,Pujades-Rodriguez, & Egger, 2009; Geng et al., 2011;Geng et al., 2013; Yu et al., 2008). Though it is clear thatsubstantial numbers of patients who discontinue carewill access care at other sites, many do so only after theirhealth has deteriorated (Hallett & Eaton, 2013), withresulting increases in morbidity and mortality (Kranzer,Govindasamy, Ford, Johnston, & Lawn, 2012). Irrespec-tive of the conditions of departure from a particular HIVclinic, the success, timeliness, and ultimate safety of con-nection to care at other sites are not well known.

© 2016 Informa UK Limited, trading as Taylor & Francis Group

CONTACT Matthew D. Hickey matt.hickey@ucsf.edu

AIDS CARE, 2016http://dx.doi.org/10.1080/09540121.2016.1179253

Dow

nloa

ded

by [

UC

SF L

ibra

ry]

at 2

3:52

04

May

201

6

This study examines outcomes among HIV-infectedpatients who stopped accessing care at a rural primaryhealth center located on an island in Lake Victoriathat serves a population in which mobile livelihoodsare particularly common (Kwena, Camlin, Shisanya,Mwanzo, & Bukusi, 2013). We examine patients wholeft this clinic with official documentation of transferas well as patients who left without documentation(i.e., LTFU). We followed these patients after clinicdeparture to determine whether they reengaged incare and the timing of any clinic-to-clinic movement.Our intention is to understand patterns of patientmovement between clinic sites, identify gaps in carethat could compromise patient safety, and explorereasons for these gaps.

Methods

Patients

This study was based at the Sena Health Center, a ruralpublic-sector clinic on Mfangano Island in Homa BayCounty, Kenya. This health center is the largest of sixfacilities on the island, serving a population of approxi-mately 21,000 (Mbita Division Population Projection,2009). All adult HIV-infected patients on combinationantiretroviral therapy (cART) were first invited to enrollin a parent study known as the Mfangano IslandHealthy Networks Impact Study (MIHNIS), a quasi-experimental trial to evaluate a social support strategyfor engagement in care (Hickey et al., 2014; Hickeyet al., 2015). Patients in the MIHNIS trial were followedfor two years, from 11/2011 to 11/2013. For the presentanalysis, we included all MIHNIS participants whotransferred-out or were LTFU at any point duringthese two years of follow-up.

Procedures and measurements

Patient demographics, travel time to clinic, perceivedHIV-related stigma in the community (Visser, Kershaw,Makin, & Forsyth, 2008), and HIV knowledge (Carey &Schroder, 2002) were measured at study enrollmentthrough surveys administered by study staff. Clinicdata, including CD4 counts, date of ART initiation,and both scheduled and attended visit dates, were cap-tured from the clinic’s OpenMRS electronic medicalrecord maintained by Family AIDS Care and EducationServices (FACES) (Kulzer et al., 2012). At the end ofstudy follow-up, patients who had left care were tracedto ascertain whether they had reengaged in HIV care.Reengagement was defined as return to the original oranother clinic to resume HIV care. For patients reportingtransfer to other facilities, chart abstraction was con-ducted at those facilities to determine dates of linkage.Patients who returned to the Sena Health Center afterinitially meeting the LTFU criteria were identifiedthrough the electronic medical record. The MIHNISstudy was approved by the KEMRI Ethical Review Com-mittee and the UCSF Committee for Human SubjectsResearch.

Analyses

We plotted the cumulative incidence of reengagement incare following the last attended visit at the Sena HealthCenter for patients who left with an official transfer ver-sus those who were LTFU. The date of last attendedappointment at the Sena Health Center was taken to betime zero for all time-to-event analyses. Cumulative inci-dence estimates and 95% confidence intervals (CIs) werecalculated using the cumulative incidence function,treating death as a competing risk (Satagopan et al.,2004). We also report the proportion of patients ineach group who experience a gap in ART adherence,based on the amount of medication dispensed at thepatient’s last clinic appointment at Sena. We evaluateddemographic and clinical factors associated with re-con-nection to care, and compared rates of reengagement,using a Cox proportional hazards model. Each predictorof interest was included in a univariable Cox model toevaluate the association between the predictor and reen-gagement in care. The multivariable model included allvariables with p < 0.1 in the univariable models.

Results

Of the 369 patients on ART at the Sena Health Center atenrollment, 23 (6%) obtained an official transfer letter toanother facility and 78 (21%) missed a clinic visit for ≥90

Figure 1. Time to re-connection to care after departure fromindex clinic.

2 M. D. HICKEY ET AL.

Dow

nloa

ded

by [

UC

SF L

ibra

ry]

at 2

3:52

04

May

201

6

days without requesting transfer. All patients were tracedto confirm whether and when they had accessed care at adifferent health facility (Table 1).

Among the 23 patients making an official transfer, 15(65%) filed for a transfer to another health facility onMfangano or neighboring Takawiri islands, while 8(35%) filed for transfer to a facility elsewhere in Kenya.Ninety-six percent reached their destination facility,with a cumulative incidence of reaching the destinationfacility by three months after their last appointment atthe Sena Health Center of 91% (95% CI 69–98%)(Figures 1 and 2). The standard practice at Sena is to dis-pense either 28 or 56 days of cART upon transfer.Among this sample, 8 (35%) were given 28 days ofcART, 8 (35%) were given 56 days of cART, and 7(30%) had missing data regarding the amount of cARTdispensed. Thus, conservatively assuming that thosewith missing data were given 28 days of therapy, 7(30%) made the transfer without any treatment gap(i.e., medication supply given by Sena exceeded timeuntil connection to a new clinic), 6 (26%) sustained atreatment interruption of 14 days or less, and 10 (43%)sustained a treatment interruption of more than 14 daysduring the transfer.

Among the 78 patients who were LTFU without offi-cial transfer (missed a clinic visit by ≥90 days), 38 (49%)eventually returned to care at the Sena Health Center, 22(28%) transferred to another facility, and 18 (23%) neverreturned to care. Two patients were unable to be locatedand were thus assumed to be out of care. The cumulativeincidence of reengagement in clinical care at a new clinicwas 14% (95% CI 7–23%) at three months and 24% (95%CI 16–34%) at six months following the last attendedappointment. The cumulative incidence of return tothe Sena Health Center was 0% at three months and36% (95% CI 26–47%) at six months (Figure 2). Thecumulative incidence of reengaging in care overall, toeither a new clinic or the original clinic, was 14% (95%CI 7–23%) at three months and 60% (95% CI 48–69%)at six months after the most recent attended clinicappointment at the Sena Health Center (Figures 1 and 2).

From the last attended appointment, the medianreturn date for the LTFU group was scheduled 56 dayslater (range 14–84 days). This corresponds to the quan-tity of drugs given to patients upon their last visit at theSena Health Center before missing a clinic visit andbecoming LTFU. The missed visit marks the time pointwhen patients likely ran out of cART, assuming thatthey were previously taking medications consistentlyand did not first transfer to another facility or obtainmedications from other non-clinic sources. Of thosewho were LTFU, 71 of 78 (91%) experienced a treatmentgap of greater than 14 days. Of the seven patients who

did not experience a treatment gap, two patients diedwithin 14 days of missing their appointment and theremaining 5 reengaged in care at another health facilitywithin 14 days of their missed appointment.

In univariable Cox proportional hazards modeling,patients who left care with an official transfer letterrelinked to care at 5.5 times the rate of thosewho left with-out a transfer letter (HR 5.5, 95% CI 3.2–9.3) (Table 2).Each one-point increase in perceived stigma in the com-munity, asmeasured by the 17-point Parallel Stigma Scale(Visser et al., 2008), was associated with an 8% relativedecrease in the rate of relinkage to care (HR 0.92, 95%CI 0.86–0.98). Having been on cART for a longer dur-ation at the time of departure from the clinic was alsoassociated with an increase in the rate of reengagement(HR 1.16 per 1-year increase, 95% CI 1.04–1.3).

The final model yielded similar effect sizes for leavingwith an official transfer, perceived stigma, and time sincecART initiation; however, only leaving with an officialtransfer (adjusted hazard ratio (aHR) 6.1, 95% CI 3.4–11.0) and perceived stigma (aHR 0.90, 95% CI 0.84–0.96) remained statistically significant.

Discussion

In our study, HIV-infected patients receiving cART wholeft their original clinic usually reconnected to care,though rates of reengagement differed based on the cir-cumstances of departure. Those who left with an officialtransfer reengaged in care at six times the rate of patientswho were LTFU. In both groups, however, treatmentinterruptions were frequent, suggesting vulnerability topoor health outcomes (Kranzer et al., 2012). The workby Parenti et al. suggests that 50% of patients who experi-ence a 15-day treatment interruption on a non-nucleo-side reverse transcriptase inhibitor (NNRTI)-basedregimen subsequently experience virologic rebound(Parienti et al., 2008). This suggests that a substantialproportion of patients in our study, where 96% ofpatients were on an NNRTI-based regimen, may alsohave experienced virologic rebound. Among patientsmaking official transfers, approximately 40% experi-enced a treatment gap greater than 14 days, implyingthat 20% of those leaving with an official transfer couldhave experienced a treatment gap-related viral rebound.The situation among those who stopped without an offi-cial transfer was worse: 91% of these patients had a treat-ment gap >14 days, implying that nearly 50% of patientsleaving clinic without an official transfer could haveexperienced viral rebound. Because many patientsexperienced gaps longer than 15 days, the proportionexperiencing viral rebound may have been even higher.In short, though neither mode of clinic departure

AIDS CARE 3

Dow

nloa

ded

by [

UC

SF L

ibra

ry]

at 2

3:52

04

May

201

6

appears to facilitate complete transition of care withoutsubstantial risk of virologic rebound, many patientstransfer care to other facilities through “unofficial” path-ways that appear to confer additional risks over thoseassociated with official transfers.

Another key finding was the association between higherdegrees of perceived stigma in the community and sub-sequent decreased rate of reengagement in care followingdeparture from the Sena clinic. This finding persisted inthemultivariatemodel and is consistent with other reportssuggesting that stigma is associated with reduced medi-cation adherence (Katz et al., 2013). Others have reportedthat patients may develop a “reluctance to return” follow-ing gaps in care (Ware et al., 2013). Our findings suggestthat stigma may play a role in exacerbating this reluctanceto return, resulting in delays in reconnection to care follow-ing a treatment interruption.

At face value, one way to interpret these data is to con-clude that all patients should obtain official transfers andthat health systems should be strict about this practice.However, we believe this oversimplified interpretation

of these data would be unfortunate for both health sys-tems and patients. In reality, many patients are engagedin livelihoods that require mobility. In communities sur-rounding Lake Victoria, one-third of households areengaged in fishing (Fiorella et al., 2014), requiring fre-quent travel to keep up with the migratory patterns offish (Camlin, Kwena, & Dworkin, 2013). Such mobilityis not always predictable. As a result, “silent transfers”,or even missed visits with subsequent return to thesame clinic, are not necessarily the result of patient fail-ure to comply with administrative requirements, butrather a reflection of the fact that livelihoods competewith health care as a priority and patients have limitedcontrol over mobility and the proximity of new or tem-porary residences to clinical services (Ware et al., 2013).We believe that health systems need to adapt to betteraccommodate patient needs regarding accessing carefrom new or multiple facilities. While transfer lettersmay improve the likelihood of successful “send-offs” tonew clinics, requiring them for intake at a “receiving”clinic would likely have serious deleterious effects for

Figure 2. Outcomes among patients leaving clinic by official transfer letter vs. those without an official transfer letter, assessed as of thedate of missed clinic appointment at the original clinic.

4 M. D. HICKEY ET AL.

Dow

nloa

ded

by [

UC

SF L

ibra

ry]

at 2

3:52

04

May

201

6

the majority of patients in our study who lacked letters,leading to longer treatment interruption or total disen-gagement from care.

Ideas for improving the coordination of transfersinclude both improving platforms for sharing clinical

data and helping patients feel more comfortable movingbetween health facilities. Sharing clinical data betweenformer and recipient health facilities can help ensurethat the patient continues to receive the same cART regi-men, without having to repeat staging or cART initiationprocedures. Such sharing platforms could include sharedelectronic medical records, patient data cards thatpatients keep with them, or a structured phone callbetween recipient and former facilities to facilitate apatient handoff. National registries of clinic contactinformation would be inexpensive to implement andcould facilitate inter-clinic communication. Proceduresto help patients feel more comfortable transferring careto another facility include more overt advertising of theprocedures by which patients can receive assistancewith transfers, as well as more friendly procedures forreceiving patients at the new facility. Clinician attitudesmay represent both a source of patient transfers and abarrier to facilitation of transfers; thus, interventions totrain clinicians to be more patient-friendly may inter-vene on multiple pathways (Ware et al., 2013).

Movement across clinic sites is fueled through threepredominant mechanisms (Figure 3). Some patientsleave their original clinics after requesting an official“transfer letter” to facilitate transition of care to another

Table 1. Participant characteristics.n or mean % or SD

Age 40 12GenderFemale 71 70%Male 30 30%

Household size 5.4 2.5Level of education completedNone 6 6%Primary 68 67%Secondary 22 22%Post-secondary 5 5%

Marital statusSingle/never married 5 5%Separated/divorced 5 5%Widowed 26 26%Married 65 64%

Monthly income (Kenyan Shillings)<2000 37 37%2000–3999 25 25%4000–5999 15 15%≥6000 23 23%

HFIAS food insecurity scale 15.3 7.0HFIAS food insecurity categoryFood secure 2 2%Mildly food insecure 2 2%Moderately food insecure 15 15%Severely food insecure 82 81%

Walking distance to Sena<30 min 56 55%30–60 min 45 45%

Most recent WHO stageI/II 48 48%III 34 34%IV 14 14%Unknown 5 5%

Baseline stigma scale 6.5 3.4Pre-transfer CD4 count 387 217Time on ART (years) 2.5 1.9Anchor drugNevirapine 89 88%Efavirenz 7 7%Lopinavir/ritonavir 5 5%

Table 2. Cox proportional hazards model: hazard of relinking tocare.

Unadjusted HR(95% CI) P

Adjusted HR(95% CI) P

Official transfer 5.51 (3.25–9.34) <.0005 6.15 (3.44–11.0)

<.0005

Age at transfer (yrs),per 10 years

1.0 (0.84–1.18) .96

GenderFemale refMale 0.71 (0.43–1.17) .18

Monthly income inKenyan Shillings

.64

<2000 ref2000–4000 1.11 (0.63–1.94)4001–6000 1.53 (0.79–2.99)>6000 1.22 (0.68–2.19)

HFIAS food insecurityscale

1.02 (0.98–1.06) .29

EducationPrimary or less RefSecondary orgreater

1.14 (0.70–1.86) .60

Attributable stigma 0.92 (0.86–0.98) .01 0.90 (0.84 to0.96)

.003

HIV knowledge 1.02 (0.91–1.14) .76Walking distancefrom Sena

.75

≤60 minutes ref>60 minutes 1.07 (0.69–1.66)Time since ARTinitiation (yrs), per 1year

1.16 (1.04 to1.30)

.01 1.08 (0.96–1.22)

.21

CD4 count at transfer .54<200 ref200–350 1.40 (0.70–2.78)>350 1.08 (0.57–2.05)

Figure 3. Three patterns of patient movement. Patients mayleave their original clinic by (1) departing for another clinicwith an official transfer letter, (2) departing for another clinicwithout an official transfer letter, and (3) ceasing care at the orig-inal clinic with or without intention of eventually returning tocare. Patients in group 3 may eventually return to care at eitherthe original clinic or another clinic.

AIDS CARE 5

Dow

nloa

ded

by [

UC

SF L

ibra

ry]

at 2

3:52

04

May

201

6

site (Nglazi et al., 2013). Other patients depart from theiroriginal site without notifying the clinic, and are thuslabeled “lost to follow-up”. This group includes bothpatients who depart with the intention of moving theircare to another site, and those who simply stop attendingappointments, with or without the intention to later returnor transfer (Brinkhof et al., 2009; Geng et al., 2010; Wareet al., 2013). Regardless of the conditions of departure,the timing of reengagement in care, at either the originalclinic or a new clinic, has important implications for thesafety and effectiveness of HIV treatment (Bastard et al.,2012;Kranzer et al., 2012;Kranzer&Ford, 2011;Mugaveroet al., 2012; Pinoges et al., 2015).

As decentralization of health facilities continues toexpand delivery of HIV care to increasing numbers ofrural sites, numbers of patients who officially transfercare and who are LTFU from their original clinic areincreasing (Fox & Rosen, 2010; Nglazi et al., 2011; Nglaziet al., 2013). Though outcomes among those who leave aparticular clinic are largely unknown, some tracingstudies have confirmed that substantial numbers ofpatients who are classified as LTFU actually made a silenttransfer to another facility (Brinkhof et al., 2009; Genget al., 2011). To date, tracing studies of patients who areLTFU have not ascertained the timing of reengagementin care following silent transfer. Two studies evaluatingtransfer outcomes following clinic-initiated “official”transfer have concluded that nearly all patients activelyreferred for transfer reach their destination clinic; how-ever, timing of transfer linkage was not assessed (Cloeteet al., 2014; O’Connor, Osih, & Jaffer, 2011). Only oneprior study has reported on the timeliness of linkage tothe new clinic following transfer, concluding that nearlyall patients making official transfers out of a large hospi-tal-based clinic inMalawi reached their destination facili-ties, with a mediation duration between departure andreengagement of 1.3 months (Yu et al., 2008). All threeof these studies, however, have focused on official trans-fers from a central site to lower level health facilities.Our study expands on this literature by describing thetimeliness of reengagement in care following departurefrom a rural primary care HIV clinic. Patients makingan official transfer in our cohort had similar successand rates of linkage to these prior studies in urban set-tings. In contrast, we observed that patients who becomeLTFU from their clinic, including those who self-transferto another clinic, frequently sustain treatment interrup-tions prior to reengaging in care.

Limitations

Though this study was small, we believe the setting isemblematic of many rural settings in East Africa: the

health facility is staffed by health workers from the Min-istry of Health, it is supported by an academic partner-ship (FACES), and the patient population served isengaged in livelihoods that are typical of the region.Overall, we believe that findings from other clinic sitesin the region are likely similar.

Second, we do not know whether patients who wereLTFU intended to go to a different facility when theyleft their original facility, or whether they intended tostop care and only reengaged when they felt unwell(Figure 3). If the former is true, this suggests that accessto care is a problem once patients leave their originalclinic. If the latter is true, it implies that patients whoare silent transfers may lack motivation or empowermentto play amore active role in their care. Althoughwe founda strong association between having a transfer letter andmore rapid reengagement among patients who stoppedattending the original clinic, we did not account for allunmeasured common causes of this relationship, thusprecluding a clear causal interpretation.We did, however,measure and adjust for several plausible confounders,including measures of socioeconomic status and stigma,and the association remained large. We therefore believethat the process of obtaining a transfer has an importanteffect on reengagement and that making transfers moreadministratively streamlined may help reduce gaps incare as patients move from one facility to another.

Further, among patients who did transfer, the reasonfor the transfer is not known. Reasons are likely multipleand may include migration, wanting to move to a moreconvenient facility, problems with the clinic staff at theold facility, or concerns about confidentiality at the oldclinic (Geng et al., 2010; Ware et al., 2013). More defini-tive information about the patient perspective wouldallow stronger inferences, and also more clearly identifyareas for intervention.

Conclusion

In conclusion, this is the first study that we know ofwhich explicitly assesses the timeliness and completenessof reengagement in care among patients who transfer orare lost to follow-up. We found that official transfersnearly always connected to the new clinic. Yet for bothofficial transfers and those who left without an officialtransfer, a substantial proportion experienced treatmentgaps during the care transition. We believe that transfers– whether silent or official – are part of the natural his-tory of treatment in resource-limited settings. Systemsmust be patient centered to optimize engagement andthereby overall effectiveness over time by making trans-fers easy and accessible. Assessing the safety of transfers,to date rarely done, needs to be a core component of

6 M. D. HICKEY ET AL.

Dow

nloa

ded

by [

UC

SF L

ibra

ry]

at 2

3:52

04

May

201

6

promoting high-quality HIV care in resource-limitedsettings.

Acknowledgements

We would like to thank the research team and staff at theEkialo Kiona Center for their tireless work. We would alsolike to thank the Director of the Centre for Microbial Researchat the Kenya Medical Research Institute (KEMRI) for supportof this project. Most importantly, we would like to thank thestudy participants in the MIHNIS trial for their willingnessto contribute time toward improving HIV care delivery inthe region. This paper was published with the permission ofthe Director, KEMRI.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was supported by Doris Duke Charitable Foun-dation; Rise Up Foundation; Horace W. Goldsmith Foun-dation; Mulago Foundation; Google Inc. via the TidesFoundation; Craigslist Foundation; Segal Family Foundation;and UCSF School of Medicine Dean’s Research Fellowship.

References

Bastard, M., Pinoges, L., Balkan, S., Szumilin, E., Ferreyra, C.,& Pujades-Rodriguez, M. (2012). Timeliness of clinicattendance is a good predictor of virological response andresistance to antiretroviral drugs in HIV-infected patients.PLoS One, 7(11), e49091. doi:10.1371/journal.pone.0049091

Brinkhof, M. W., Pujades-Rodriguez, M., & Egger, M. (2009).Mortality of patients lost to follow-up in antiretroviral treat-ment programmes in resource-limited settings: Systematicreview and meta-analysis. PLoS One, 4(6), e5790. doi:10.1371/journal.pone.0005790

Camlin, C. S., Kwena, Z. A., & Dworkin, S. L. (2013). Jaboya vs.jakambi: Status, negotiation, and HIV risks among femalemigrants in the “sex for fish” economy in NyanzaProvince, Kenya. AIDS Education and Prevention, 25(3),216–231. doi:10.1521/aeap.2013.25.3.216

Camlin, C. S., Snow, R. C., & Hosegood, V. (2014). Genderedpatterns of migration in rural South Africa. Population,Space and Place, 20, 528–551. doi:10.1002/psp.1794

Carey, M. P., & Schroder, K. E. (2002). Development and psy-chometric evaluation of the brief HIV knowledge question-naire. AIDS Education and Prevention, 14(2), 172–182.

Cloete, C., Regan, S., Giddy, J., Govender, T., Erlwanger, A.,Gaynes, M. R.,… Bassett, I. V. (2014). The linkage out-comes of a large-scale, rapid transfer of HIV-infectedpatients from hospital-based to community-based clinicsin South Africa. Open Forum Infectious Diseases, 1(2),ofu058. doi:10.1093/ofid/ofu058

Fiorella, K. J., Hickey, M. D., Salmen, C. R., Nagata, J. M.,Mattah, B., Magerenge, R.,… Fernald, L. H. (2014).Fishing for food? Analyzing links between fishing liveli-hoods and food security around Lake Victoria, Kenya.

Food Security, 6(6), 851–860. doi:10.1007/s12571-014-0393-x

Fox, M. P., & Rosen, S. (2010). Patient retention in antiretro-viral therapy programs up to three years on treatment insub-Saharan Africa, 2007-2009: Systematic review.Tropical Medicine & International Health, 15(Suppl. 1), 1–15. doi:10.1111/j.1365-3156.2010.02508.x

Geng, E. H., Bangsberg, D. R., Musinguzi, N., Emenyonu, N.,Bwana, M. B., Yiannoutsos, C. T.,…Martin, J. N. (2010).Understanding reasons for and outcomes of patients lostto follow-up in antiretroviral therapy programs in Africathrough a sampling-based approach. JAIDS Journal ofAcquired Immune Deficiency Syndromes, 53, 405–411.doi:10.1097/QAI.0b013e3181b843f0

Geng, E. H., Bwana, M. B., Muyindike, W., Glidden, D. V.,Bangsberg, D. R., Neilands, T. B.,…Martin, J. N. (2013).Failure to initiate antiretroviral therapy, loss to follow-upand mortality among HIV-infected patients during thepre-ART period in Uganda. JAIDS Journal of AcquiredImmune Deficiency Syndromes. doi:10.1097/QAI.0b013e31828af5a6

Geng, E. H., Glidden, D. V., Bwana, M. B., Musinguzi, N.,Emenyonu, N., Muyindike, W.,…Martin, J. N. (2011).Retention in care and connection to care among HIV-infected patients on antiretroviral therapy in Africa:Estimation via a sampling-based approach. PloS one, 6,e21797. doi:10.1371/journal.pone.0021797

Hallett, T. B., & Eaton, J. W. (2013). A side door into care cas-cade for HIV-infected patients? JAIDS Journal of AcquiredImmune Deficiency Syndromes, 63, S228–S232. doi:10.1097/QAI.1090b1013e318298721b.

Hickey, M. D., Salmen, C. R., Omollo, D., Mattah, B., Geng, E.H., Bacchetti, P.,… Cohen, C. R. (2015). Pulling the net-work together: Quasi-experimental trial of a patient-definedsupport network intervention for promoting engagement inHIV care and medication adherence on Mfangano Island,Kenya. JAIDS Journal of Acquired Immune DeficiencySyndromes, 69(4), e127–e134.

Hickey, M. D., Salmen, C. R., Tessler, R. A., Omollo, D.,Bacchetti, P., Magerenge, R.,…Gandhi, M. (2014).Antiretroviral concentrations in small hair samples as afeasible marker of adherence in rural Kenya. JAIDSJournal of Acquired Immune Deficiency Syndromes, 66(3),311–315. doi:10.1097/qai.0000000000000154

Katz, I. T., Ryu, A. E., Onuegbu, A. G., Psaros, C., Weiser, S. D.,Bangsberg, D. R., & Tsai, A. C. (2013). Impact of HIV-related stigma on treatment adherence: Systematic reviewand meta-synthesis. J Int AIDS Soc, 16(3 Suppl. 2), 18640.doi:10.7448/ias.16.3.18640

Kranzer, K., & Ford, N. (2011). Unstructured treatment inter-ruption of antiretroviral therapy in clinical practice: A sys-tematic review. Tropical Medicine & International Health,16(10), 1297–1313. doi:10.1111/j.1365-3156.2011.02828.x

Kranzer, K., Govindasamy, D., Ford, N., Johnston, V., & Lawn,S. D. (2012). Quantifying and addressing losses along thecontinuum of care for people living with HIV infection insub-Saharan Africa: A systematic review. Journal of theInternational AIDS Society, 15(2), 17383. doi:10.7448/ias.15.2.17383

Kulzer, J. L., Penner, J. A., Marima, R., Oyaro, P., Oyanga, A.O., Shade, S. B.,… Cohen, C. R. (2012). Family model ofHIV care and treatment: A retrospective study in Kenya.

AIDS CARE 7

Dow

nloa

ded

by [

UC

SF L

ibra

ry]

at 2

3:52

04

May

201

6

Journal of the International AIDS Society, 15(1), 8. doi:10.1186/1758-2652-15-8

Kwena, Z. A., Camlin, C. S., Shisanya, C. A., Mwanzo, I., &Bukusi, E. A. (2013). Short-term mobility and the risk ofHIV infection among married couples in the fishing com-munities along Lake Victoria, Kenya. PLoS One, 8(1),e54523. doi:10.1371/journal.pone.0054523

Marson, K. G., Tapia, K., Kohler, P., McGrath, C. J., John-Stewart, G. C., Richardson, B. A.,… Chung, M. H. (2013).Male, mobile, and moneyed: Loss to follow-up vs. transferof care in an urban African antiretroviral treatment clinic.PLoSONE, 8(10), e78900. doi:10.1371/journal.pone.0078900

Mbita Division Population Projection. (2009). Kenya:Government of Kenya.

Mugavero, M. J., Westfall, A. O., Zinski, A., Davila, J.,Drainoni, M. L., Gardner, L. I.,…Giordano, T. P. (2012).Measuring retention in HIV care: The elusive gold standard.JAIDS Journal of Acquired Immune Deficiency Syndromes,61(5), 574–580. doi:10.1097/QAI.0b013e318273762f

Nglazi, M. D., Kaplan, R., Orrell, C., Myer, L., Wood, R.,Bekker, L. G., & Lawn, S. D. (2013). Increasing transfers-out from an antiretroviral treatment service in SouthAfrica: Patient characteristics and rates of virological non-suppression. PLoS One, 8(3), e57907. doi:10.1371/journal.pone.0057907

Nglazi, M. D., Lawn, S. D., Kaplan, R., Kranzer, K., Orrell, C.,Wood, R., & Bekker, L. G. (2011). Changes in programmaticoutcomes during 7 years of scale-up at a community-basedantiretroviral treatment service in South Africa. JAIDSJournal of Acquired Immune Deficiency Syndromes, 56(1),e1–8. doi:10.1097/QAI.0b013e3181ff0bdc

O’Connor, C., Osih, R., & Jaffer, A. (2011). Loss to follow-up ofstable antiretroviral therapy patients in a decentralizeddown-referral model of care in Johannesburg, South

Africa. JAIDS Journal of Acquired Immune DeficiencySyndromes, 58(4), 429–432. doi:10.1097/QAI.0b013e318230d507

Parienti, J. J., Das-Douglas, M., Massari, V., Guzman, D.,Deeks, S. G., Verdon, R., & Bangsberg, D. R. (2008). Notall missed doses are the same: Sustained NNRTI treatmentinterruptions predict HIV rebound at low-to-moderateadherence levels. PLoS ONE, 3(7), e2783. doi:10.1371/journal.pone.0002783

Pinoges, L., Schramm, B., Poulet, E., Balkan, S., Szumilin, E.,Ferreyra, C., & Pujades-Rodriguez, M. (2015). Risk factorsand mortality associated with resistance to first line antire-troviral therapy: Multicentric cross-sectional and longitudi-nal analyses. JAIDS Journal of Acquired Immune DeficiencySyndromes. doi:10.1097/qai.0000000000000513

Satagopan, J. M., Ben-Porat, L., Berwick, M., Robson, M.,Kutler, D., & Auerbach, A. D. (2004). A note on competingrisks in survival data analysis. British Journal of Cancer, 91(7), 1229–1235. doi:10.1038/sj.bjc.6602102

Visser, M. J., Kershaw, T., Makin, J. D., & Forsyth, B. W.(2008). Development of parallel scales to measure HIV-related stigma. AIDS and Behavior, 12(5), 759–771. doi:10.1007/s10461-008-9363-7

Ware, N. C., Wyatt, M. A., Geng, E. H., Kaaya, S. F., Agbaji, O.O., Muyindike, W. R.,…Agaba, P. A. (2013). Toward anunderstanding of disengagement from HIV treatment andcare in sub-Saharan Africa: A qualitative study. PLoSMedicine, 10(1), e1001369. doi:10.1371/journal.pmed.1001369

Yu, J. K., Tok, T. S., Tsai, J. J., Chang, W. S., Dzimadzi, R. K.,Yen, P. H.,…Harries, A. D. (2008). What happens topatients on antiretroviral therapy who transfer out toanother facility? PLoS One, 3(4), e2065. doi:10.1371/journal.pone.0002065

8 M. D. HICKEY ET AL.

Dow

nloa

ded

by [

UC

SF L

ibra

ry]

at 2

3:52

04

May

201

6