Subscribe to E-Bulletin Donate to CEGA

The Effects of Nurse Absenteeism on Health Outcomes for Pregnant Women with HIV

Development Challenge

Absenteeism is common among public service workers, particularly in education and healthcare, throughout the developing world; evidence suggests that absentee rates are as high as 40 percent in some countries.[1]  Absenteeism affects public service delivery and overall quality of care. Health provider absences are compounded by shortcomings in service delivery processes and patient tracking procedures. While there is some evidence on the negative effects of teacher absenteeism on learning, little research has quantified these effects on health outcomes.


In 2007 Nyanza Province had the highest HIV prevalence rate in Kenya – 17.2 percent among women[2] –and preventing mother to child transmission (PMTCT) through antenatal care (ANC) visits was a high public health priority. This study was conducted in a rural ANC clinic in the Maseno Division of Nyanza Province. While absenteeism rates there were relatively low, there was only one nurse trained in HIV testing and counseling, so her presence was crucial for PMTCT. Women who visit the clinic typically make three to four visits. They are often only offered HIV testing during their first, although they can request to get tested at subsequent visits. If a woman tests HIV positive, she receives additional PMTCT services.

Evaluation Strategy

This study sought to measure the impacts of nurse absence on HIV testing, receipt of PMTCT services, and other maternal and child healthcare outcomes. Researchers collected data on 591 women from 2005-2007 using both in-clinic and household surveys. Additional administrative data collected included daily attendance of the PMTCT nurse, women’s consent to the HIV test, and the test result itself (with patient consent).

To identify the impact of nurse absence on health outcomes, researchers used reduced form and cross-sectional regressions controlling for several possible biasing factors. The study also analyzed the differential impact of nurse presence on women with and without HIV, as women who are HIV positive would experience more negative consequences from nurse absence. Because the actual HIV status was not known for all women, self-reported beliefs were used as a proxy measure for actual HIV status. Women who believed they had a low risk were categorized as low prior while women who believed they had a moderate or great risk were considered high prior.

The primary health outcomes measured included:

  • HIV testing;
  • Delivery in a hospital or health clinic;
  • Number of ANC visits;
  • Medication to prevent mother to child HIV transmission;
  • Breastfeeding; and
  • Enrollment in the AMPATH treatment program.

Results and Policy Implications

The results indicate that nurse absenteeism negatively affects pregnant women’s health outcomes. Specifically, if a nurse was present, a woman was 55 to 59 percentage points more likely to be tested for HIV. This means a woman whose first visit coincides with the nurse’s attendance is three times more likely to learn her HIV status, while only one out of four women is likely to learn their HIV status in subsequent visits. This suggests the referral system is broken, as poor record keeping fails to identify women still in need of HIV testing and counseling.

If a nurse was present during a patient’s first visit, she was 13 percentage points more likely to deliver at a health center. This corresponds to an almost fifty percent increase in the probability that a woman delivers in a facility. Additionally, depending on whether women were classified as low or high prior, nurse presence affected the likelihood that they took PMTCT medication, breastfeed, or used the PMTCT services. The effects are detailed in the table below.

Effect of Nurse Absenteeism on Health Outcomes

Change in percentage points when nurse was present

Health Outcome

Among High-prior women

Among Low-prior women

Choice to deliver in hospital or health center



Report receiving PMTCT medication


No effect

Breast-feeding patterns


No effect

Enrollment in AMPATH treatment program


No effect

These results suggest that the PMTCT nurse’s presence affects pregnant women’s behavior, and their presence during the first visit has larger, positive gains for HIV-positive or high risk women. In fact, the data suggests that the nurse’s absenteeism rate – 9 percent – contributed to an additional 3.7 mother-to-child transmissions per 10,000 live births. While this study only includes one clinic in rural Kenya, the conditions of the study site – provider absenteeism and ineffective or lack of follow-up – are common across many sub-Saharan African countries. Reducing health worker absenteeism at ANC clinics or improving follow-up procedures may be an effective strategy to reduce the number of new HIV cases by reducing mother-to-child transmission.



[1] Chaudhury, Nazmul, Jeffrey Hammer, Michael Kremer, Karthik Muralidharan, and F. Halsey Rogers. 2006. “Missing in Action: Teacher and Health Worker Absence in Developing Countries.” Journal of Economic Perspectives 20 (1): 91–116. 

[2] National AIDS, and STI Control Program. 2009. “Kenya AIDS Indicator Survey (KAIS) 2007 Data Sheet.” Nairobi: National AIDS and STI Control Program. 

3. Photo Credit: Moving Mountains Trust, Clinic in Western Kenya, 2010.