Pharmacy staff also critical for keeping ART patients in care, Mozambique study suggests

Pharmacy staff may be just as important as clinical staff in ensuring that patients in sub-Saharan Africa are not lost to follow-up, according to a large study of public sector HIV care and treatment clinics in Central Mozambique.

Higher pharmacy staff burden was an important predictor of loss-to-follow-up or death among 12,000 patients starting antiretrovirals, Barrot H Lambdin and colleagues reported in a retrospective cohort study published in the advance online edition of the Journal of Acquired Immune Deficiency Syndromes.

There was no significant association between clinical staff burden and attrition. The authors suggest that task-shifting partly explained this finding. Additionally, clinical monitoring needed fewer visits than the required monthly pharmaceutical visits.

While patients attending high-volume clinics were at a higher risk for loss-to-follow up or death than those attending low-volume clinics, this difference was not statistically significant (p=0.198).

Seven thousand people were on ART in Mozambique in 2004. This number increased dramatically to 170,000 by the end of 2009.

Mozambique has a critical shortage of health care personnel with one of the lowest provider-to-population ratios in the world: three doctors, 21 nurses and three pharmacy staff for every 100,000 people.

Inadequate human resource capacity is considered to be one of the most significant barriers to effective ART delivery in resource-poor settings.

Increasing patient populations can potentially overwhelm an already overstretched staff and lead to bottlenecks within the system. So patients would spend more time waiting for less quality time with their provider leading to patient dissatisfaction and ultimately higher loss-to follow-up.

The authors chose to look at how patient volume, health workforce levels (clinical and pharmacy staff) and patient characteristics affected retention in ART programmes in Manica and Sofala provinces in Central Mozambique. Understanding the effect, they note, is critical to ensure effective delivery of HIV care and treatment.

HIV prevalence in Central Mozambique is amongst the highest in the country, at 20.4% in 2004. The first two ART delivery sites were set up in 2003 and 2004 at Sofala’s Beira Central Hospital and Manica’s Chimoio Provincial Hospital, respectively.

In 2006 HIV services were integrated into primary health care clinics across the two provinces’ 23 districts with the aim of improving access to HIV care and treatment.

77% (11,793) of those starting ART in the study period were included in the analysis. Those excluded were under 15 years of age, pregnant or had transferred to another facility.

At the end of the study period 63% (7,491) patients were alive and on ART from the clinics where they started; 16% (1,932) were lost to follow-up and 14% (1,645) died.

There were considerable differences among clinics in the number of months providing treatment, the number of adults starting treatment, patient volume and human resource levels (clinician and pharmacy staff burden).

Patient volume is defined as the monthly number of clinical and pharmacy visits; clinical staff burden as the monthly number of clinical visits for each clinician and pharmacy staff burden the monthly number of pharmacy visits for each pharmacy staff person.

The median number of monthly visits at the clinics was 562 (IQR: 264-1,141). The average number of clinicians providing care each month was 2.89; and the mean number of pharmacy staff providing services each month was 1.21.

So the median monthly number of clinic visits for each clinician was 111 (IQR:61-214); and the median number of pharmacy visits for each pharmacy staff person was 359 (IQR: 142-609).

After adjusting for patient characteristics patients attending clinics with medium pharmacy staff burden HR=1.39 (95% CI: 1.07-1.80) and high pharmacy staff burden HR= 2.09 (95% CI: 1.50-2.91) tended to have a higher risk for attrition than those with low pharmacy staff burden (p <0.001).

Even after adjusting for a year on treatment, clinic location and clinic experience the association between pharmacy staff burden and attrition became stronger.

Strengths of the study include standardisation of protocols for delivery of care, patient tracing and data recording across clinics, note the authors.

The main limitation, they add, is the observational nature of the study, raising the possibility that unmeasured patient and clinic factors could bias the results.

Caution is needed, the authors add, before generalising these results to other settings. Clinics were selected if they had an electronic database, not through a random process.

Attrition at 12 and 24 months of 32% and 41% respectively, the authors suggest was also probably due to over 65% of patients starting ART late (WHO Stages III and IV) and health system bottlenecks.

Nevertheless the authors conclude “pharmacy staff burden was an important predictor of attrition” highlighting “a potential area within the health system where interventions could be applied to improve retention.”