About

Katherine (Kat) Tumlinson, PhD, is an Associate Professor of Health Policy and Management at the UNC Gillings School of Global Public Health and a Fellow of the Carolina Population Center. She is trained as an epidemiologist and demographer, with a focus on the intersection of population and global reproductive health. Dr. Tumlinson has served as a principal investigator on research studies conducted in the Global South, involving cross-sectional and longitudinal primary data collection at both household and healthcare facility levels. Her professional background includes work as a family planning service provider, reproductive health educator, and advocate, which informs her highly relevant and applied research agenda. Her research primarily concentrates on global reproductive justice, emphasizing increasing access to high-quality contraceptive care in low- and middle-income countries. Notably, her recent work in Western Kenya identified significant health system barriers to contraceptive access, highlighting the importance of addressing provider motivation and systemic issues to improve reproductive health outcomes.

Research topics

• Medicine
• Political Science
• Environmental health
• Nursing
• Psychology
• Sociology
• Family medicine
• Economic growth
• Pediatrics
• Business
• Economics
• Finance
• Demography

Selected publications

• Provider confidence & willingness to insert & remove long-acting reversible contraception methods in Kenyan public facilities

  UNC Libraries · 2026-05-06

  articleOpen access
  PublisherDOI

• Reasons for contraceptive nonuse at two sites in Burkina Faso

  Contraception · 2026-03-01

  articleOpen accessSenior author

  OBJECTIVE: We examine reasons for contraceptive nonuse in Burkina Faso. STUDY DESIGN: We draw from a population-based survey of 3929 women, of whom 2462 were nonusers of contraception. We use descriptive statistics to examine the distribution of reasons for nonuse. RESULTS: Over 80% of respondents reported lack of desire or need for contraception as motivating nonuse. Conventional access barriers, like cost or distance, were uncommon. CONCLUSIONS: Restricted access to contraception was not a commonly reported barrier to contraceptive use in this context. IMPLICATIONS: Sexual and reproductive health researchers should avoid assuming that nonusers lack contraceptive access. Future research should explore further reasons for contraceptive nonuse.

  PublisherOA PDFDOI

• Postpartum Acute Care Utilization in a Health Care System in the Southeastern United States

  UNC Libraries · 2026-05-06

  articleOpen access1st authorCorresponding

  <strong><em>Introduction:</em></strong> Postpartum acute care utilization (PACU), including visits to an emergency department, obstetric triage, or urgent care ("outpatient"), and hospital readmissions, may indicate medical complications and signal unmet health needs. <strong><em>Methods:</em></strong> We estimated the incidence of PACU and examined patterns by sociodemographic factors, pregnancy and birth characteristics, time since discharge from the birth hospitalization, and medical indications. We constructed a retrospective cohort of people aged &ge;18 years who delivered &ge;1 liveborn infant &gt;20 weeks of gestation from July 1, 2021, to December 31, 2022, using electronic health record data from a quaternary maternity hospital in the Southeastern United States PACU data throughout the health care system were collected through March 31, 2023. We excluded people with a hospital stay &gt;6 days (<em>n</em> = 29). <strong><em>Results:</em></strong> In this cohort of 6,041 birthing people, 11.3% had &ge;1 outpatient encounters&nbsp;(range 0-6) and 3.2% had &ge;1 hospital readmissions (range 0-4) within 12 weeks of discharge from the birth hospitalization. Median time to first outpatient PACU was 10 days post-discharge and 6 days for first hospital readmission. Among encounters for the top five medical indications, time to first postpartum acute care encounter varied by medical indication (log-rank test of equality over strata Chi-square = 69.93, degrees of freedom = 4, <em>p</em> &lt; 0.0001). Complications specified during the puerperium (<em>n</em> = 234) and hypertension and hypertensive-related conditions complicating the puerperium (<em>n</em> = 87) were the two most frequent indications. <strong><em>Conclusion:</em></strong> These findings can inform efforts to direct health resources to improve postpartum health care and health outcomes.

  PublisherDOI

• SAFI-FP Baseline Unannounced Visit User Guide

  UNC Libraries · 2026-01-08

  articleOpen accessSenior author

  Data were collected as part of the baseline for a cluster randomized controlled trial in Kisumu County, Western Kenya, evaluating a social accountability&ndash;based quality improvement intervention designed to strengthen contraceptive service delivery through citizen feedback, public monitoring, and dialogue. Baseline data were gathered from all 137 public-sector facilities between May and October 2022 using structured unannounced visits and provider interviews. A very short facility audit was also conducted in between unannounced visits and provider interviews.

  PublisherDOI

• Keeping the "R" in LARC (long-acting reversible contraception): Measuring client-centered implant removal services in sub-Saharan Africa.

  UNC Libraries · 2025-08-02

  articleOpen access

  As the use of subdermal contraceptive implants increases across sub-Saharan Africa, the need for person-centered removal services is more critical than ever to safeguard reproductive autonomy. In 2016, Christofield and Lacoste proposed eight conditions for client-centered implant removal, yet the extent to which these conditions have been assessed in large-scale surveys remains unexamined. Our mapping exercise collates survey information from three large data collection platforms fielded in sub-Saharan Africa, including the Demographic and Health Surveys (DHS), Performance Monitoring for Action (PMA), and the Service Availability and Readiness Assessment (SARA), utilizing questionnaire tools implemented among women, health facilities, providers, and clients to map existing data sources against these conditions. Our findings reveal that four of the eight conditions are fully captured, three are partially captured, and one is entirely absent within current population or facility-based surveys. Specifically, among the six facility-based conditions: the three conditions completely captured include the availability of supplies (condition 2), systems for managing difficult removals (condition 3), and the provision of counseling and reinsertion options (condition 5); two conditions&nbsp;partially captured include competent and confident providers (condition 1) and affordability for clients (condition 7), and the one condition that remained&nbsp;absent was measurement systems for monitoring of removal data (condition 8). Among the two individual-level conditions, timely and proximate service (condition 4) was partially captured and knowledge and awareness of where to go for removal (condition 6) was fully captured. Nearly a decade after Christofield and Lacoste proposed these client-centered conditions, the lack of consistent measures for client-centered implant removal poses significant barriers to understanding service accessibility and women's reproductive experiences. To protect the reproductive autonomy of individuals who use long-acting reversible contraceptive (LARC)&nbsp;methods and desire timely discontinuation, it is imperative to develop and implement standardized metrics for their removal.

  PublisherDOI

• Measuring misaligned contraceptive use among reproductive-aged women in Kisumu, Kenya: a cross-sectional, population-based study

  BMJ Public Health · 2025-03-01 · 4 citations

  articleOpen accessSenior author

  Introduction: Global family planning scholars have critiqued traditional measures of programmatic success and called for new, person-centred measures that consider people's preferences. We propose a new measure that assesses the alignment between an individual's contraceptive desires and use. Methods: We use data from a population-based survey implemented among adult reproductive-aged women in Kisumu, Kenya. We define misaligned contraceptive use as discordance between contraceptive desires and use (ie, desire for contraception paired with non-use or no desire for contraception paired with use). We expand misaligned use to incorporate non-preferred contraceptive method use, defined as the use of a method with a preference for a different method among contraceptive users. We also compare unmet need for family planning with misaligned use. Results: In our total sample, including contraceptive users and non-users, 86% of participants had aligned contraceptive use and 14% had misaligned use. Among contraceptive users, we found that 74% of participants had aligned and preferred method use, 21% had aligned and non-preferred method use, 2% had misaligned and preferred use, and 3% had misaligned and non-preferred use. Our comparison of misaligned use and unmet need showed little agreement. Conclusions: Measuring misaligned and non-preferred contraceptive method use has implications for family planning measurement and programming. Our results demonstrate the importance of capturing contraceptive desires rather than assuming an implicit desire for contraception among those who do not desire pregnancy. Further, incorporating method preferences into misaligned use provides additional detail on who does and does not have their contraceptive desires met. Finally, measuring misaligned and non-preferred method use highlights subsets of populations who may benefit from additional contraceptive programming to improve contraceptive access or enable method discontinuation or switching, and therefore can be imperative to ensuring that individuals have their reproductive desires realised.

  PublisherOA PDFDOI

• Measuring Informed Choice for Contraception in Burkina Faso: Comparing Self-Rated and Researcher-Ascribed Measures

  Social Indicators Research · 2025-07-02

  article
  PublisherDOI

• Provider confidence and willingness to insert and remove long-acting reversible contraception methods in Kenyan public facilities

  Utrecht University Repository (Utrecht University) · 2025-09-01

  articleOpen access1st authorCorresponding

  Objective: Use of long-acting reversible contraception (LARC) in Kenya has grown over the last decade, yet emerging evidence points to challenges in LARC removal. The objective of this paper is to document provider training in LARC insertion/removal and to better understand provider experience, confidence, and willingness to both insert and remove LARC. Study design: In this paper we present a descriptive analysis of self-reported data from family planning providers working in all 137 public-sector healthcare facilities in Kisumu County (Western Kenya). We assess the frequency of public-sector provider training, experience, confidence, and willingness to insert and remove LARC, as well as the association between confidence and willingness to provide these services. Results: A substantial proportion of providers in our study lacked training in the insertion and removal of both implants and intrauterine devices (IUDs). Overall, providers reported that they were more experienced, confident, and willing to insert implants rather than remove them. A fifth of providers trained in IUD insertion were not confident in their ability to insert IUDs and nearly 10% were not willing to perform insertion. In contrast, 8% of providers trained in IUD insertion were not confident in their ability to remove IUDs and 6% were not willing to remove them. Provider confidence was significantly associated with willingness to insert/remove both implant and IUD. Less than two-thirds of facilities demonstrated stock of the implant (61%) or IUD (58%). Conclusion: Results raise important concerns about reproductive autonomy within the larger environment of family planning programs that heavily promote LARC use. Within Kisumu, public-sector providers require greater supplies and support to maintain confidence and ensure their willingness to engage in LARC insertion and removal in equal measure. Implications: We find that many providers at public sector healthcare facilities in Kisumu, Kenya lacked training in LARC insertion and removal. Providers reported that they were more experienced, confident, and willing to insert rather than remove implants, which poses a potential threat to reproductive autonomy.

  Publisher

• Postpartum Intrauterine Device Removal and Access to Removal in the 18 Months Following an Intervention in Tanzania, Sri Lanka, and Nepal

  Studies in Family Planning · 2025-10-23

  articleOpen access

  Family planning programs in the Global South promote long-acting reversible contraception (LARC), but research suggests that women face barriers to LARC discontinuation, inhibiting their reproductive autonomy. Scholars have called for improved data visibility around LARC removal access. We use data from the Postpartum Intrauterine Device (PPIUD) Study, a randomized trial of a PPIUD intervention conducted in Nepal, Sri Lanka, and Tanzania from 2015 to 2018. Among women who adopted PPIUDs, we describe PPIUD status (in-use, expelled, deliberately removed) at three follow-up points. We report the proportion of participants who sought PPIUD removal and, among those, whether they faced barriers to removal. About three-quarters of 5370 participants had their PPIUD in use 18 months following insertion; one-fifth had their PPIUD deliberately removed, and 6 percent had their PPIUD expelled. Of the 22 percent of participants who sought PPIUD removal, a quarter faced a barrier to removal; most barriers were provider-imposed. In conjunction with existing literature, our findings highlight that barriers to LARC removal are structural, rooted in larger public health and demographic goals that aim to promote contraceptive uptake. We call for safeguards to ensure that people who desire LARC removal can discontinue their method.

  PublisherOA PDFDOI

• Barriers to contraceptive implant removal among current and prior implant users in Kisumu, Kenya: Results from a cross-sectional population-based survey.

  UNC Libraries · 2025-11-21

  articleOpen access

  There has been increased promotion of long-acting reversible contraception (LARC) in the Global South in recent years. Studies have documented that women face barriers to removal when they desire to discontinue their LARC method, though few have quantified how often such experiences occur. We estimated the proportion of current and prior implant users who experienced or anticipated challenges with method removal using secondary data from a population-based survey administered to reproductive-aged women (18-49 years) in Kisumu, Kenya. We included women who reported current contraceptive implant use (n&thinsp;=&thinsp;518) or ever using an implant that was later removed (n&thinsp;=&thinsp;838). We asked current implant users about their anticipated barriers to removal and experiences seeking removal. We asked those who reported prior implant removal about barriers they faced when seeking removal. Around 40% of current implant users anticipated barriers to implant removal. About 2% of current implant users had sought removal unsuccessfully in the last year, with cost being the most common barrier. Of those who had ever had an implant removed, 15% reported facing challenges with removal, with cost again being the most common barrier. Sociodemographic characteristics were not associated with facing challenges with removal. Many women anticipated or experienced barriers to implant removal. Cost was the most common barrier reported, highlighting the importance of eliminating fees associated with removal to ensure all women can discontinue their method when desired.

  PublisherDOI

Recent grants

• Service Delivery Factors Influencing Contraceptive Use Dynamics in Developing Countries: A Mixed-Methods and Cross-Disciplinary Approach

  NIH · $665k · 2017–2021

• Service Delivery Factors Influencing Contraceptive Use Dynamics in Developing Countries: A Mixed-Methods and Cross-Disciplinary Approach

  NIH · $241k · 2016–2018

Frequent coauthors

• Dickens Onyango

  Ministry of Health

  24 shared

• Leigh Senderowicz

  University of Wisconsin–Madison

  20 shared

• Brooke W. Bullington

  University of North Carolina at Chapel Hill

  15 shared

• Caitlin R. Williams

  15 shared

• Laura E. Britton

  9 shared

• Debborah Muthoki Wambua

  Practical Action

  9 shared

• Siân Curtis

  St Michael’s Hospital

  8 shared

• Ilene S. Speizer

  7 shared

Labs

• UNC Gillings School of Global Public Health - Department of Health Policy and ManagementPI

Awards & honors

• Global Health Research Award 2021
• IBM Junior Faculty Development Award 2020
• Quality Innovation Challenge 2018
• David and Lucile Packard Online Quality Innovation Challenge…
• David and Lucile Packard Foundation Fellow 2016-2018