The integration of mother and child health care into one-stop consultations was found to be feasible and acceptable to the personnel of the selected rural public healthcare facilities in Tete Province, Mozambique. Our study shows that MCH nurses in the participating healthcare facilities were very well capable of changing their daily practice to accommodate such integration, and that they felt that this streamlined their service delivery. Integration of routine data collection tools in addition to the integrated clinical service delivery enhanced this perception of increased efficiency, also in the healthcare facilities in the control group. However, the practice of registering all consultations for follow-up of HIV-exposed infants in the same facility-based registration books rather than in individual files or patient-held records, made the actual implementation of integrated MCH care more difficult in the healthcare facilities with a larger patient flow, where more than one nurse attends clients simultaneously. Due to limitations in the number of available attendance rooms in many health facilities in Mozambique, it is not uncommon that two patients are attended by two providers in the same room at the same time, separated by a curtain or a screen, but this is obviously in conflict with the patient’s right to privacy.
In our study we did not assess client satisfaction, due to the limited local capacity to rigorously implement more complex study designs. Indirectly, the participating MCH nurses reported not to have received any complaints from clients, while in one of the intervention facilities clients apparently were pleased with the more streamlined attendance. However, before even considering any widespread introduction of one-stop, integrated MCH consultations, it would be important to assess its acceptability to the women attending these services.
Although there was high satisfaction among healthcare providers with the one-stop approach and a subjective feeling of increased effectiveness, this was not supported by the monthly MCH services attendance statistics. Attendance at most services was lower in the post-intervention period in both control and intervention facilities. The reason for this reduction in attendance is not clear, and it might have occurred due to dynamics not measured in our study. Possibly some contributing factors, such as absence of personnel, were more frequent in the post-intervention period, but data from the pre-intervention period were unfortunately not available. Another possible explanation is a higher quality of data registration in the post-intervention period, related to the integrated registration book introduced with the study in the six participating health care facilities, combined with possible over-reporting in the pre-intervention period. Use of routine data for scientific evaluations always has its limitations, in view of the lack of control over the quality of registered data. At the time of the study the MCH program in Mozambique used a multitude of different registration tools, some in books and some on loose sheets, with an irregular supply and limited availability and archiving at facility level. The rather short period for which we could collate pre-intervention data for our study (six months) is very much related to this limitation, as too often information was not traceable for months further in the past.
No pre-intervention data were available on the number of HIV-exposed infants followed at the facilities, and we could only assess potential post-intervention time trends. There were clear positive changes over time during the post-intervention period in the age at which children started follow-up, the number of visits they attended, and the age at which they were (serologically) tested for HIV. These effects were similar in the control and intervention healthcare facilities and could therefore not be attributed to the one-stop approach. In fact, our data show that all participating facilities gradually improved their follow-up of HIV-exposed infants. The earlier enrolment appears to be due to the fact that MCH nurses started to take advantage of post-natal visits to enrol the baby in the follow-up program, or even enrolled it already at birth. The number of follow-up visits of HIV-exposed infants increased as these were more and more often combined with child health visits for growth monitoring and vaccinations. Similarly, such child health visits increasingly included HIV testing for infant diagnosis of HIV-exposed infants. It is likely that these changes in daily practice of MCH nurses in both control and intervention facilities were induced by the regular supervisory visits, during which all MCH personnel received coaching and mentoring to improve their service delivery skills, including PMTCT. These quantitative data confirm the subjective sense of improved performance gained from the qualitative information from the staff interviews.
During the post-intervention period we recorded any periods that the healthcare facilities were confronted with absences of staff or with stock-outs of essential HIV testing commodities or MCH drugs. These periods were frequent, with about half of the observed months presenting stock-outs, and one in every six months presenting staff absences. Months with stock-outs had a lower number of activities involving the absent commodities and months with absence of staff had a substantial lower number of activities concerning follow-up of HIV-exposed infants. We therefore believe that these factors, combined with the regular staff support, had a greater effect on the performance of MCH services than the manner of organisation of these services.
Regular constructive supervision, adequate staffing levels and sufficient supply of commodities form important components of decent working conditions for health workers, essential to improve health worker retention and quality of services [12–14]. Unfortunately, as in the facilities in our study, such conditions are lacking in many healthcare facilities in Sub Saharan Africa. There is a severe lack of staff including midwives in many African countries, including Mozambique, which would need to double its number of midwives to remedy shortages, in addition to strengthening skills of the existing workforce [10, 15]. UNFPA in Mozambique recently reported a poor availability of drugs for contraception and obstetric care, especially in peripheral health facilities, with frequent stock-outs . WHO has also highlighted the poor availability in the public sector of a basket of essential medicines in many regions, including in Africa . Both reports recommend reinforcing the acquisition and distribution chain in general, and of medicine policies in particular. Our study suggests, both qualitatively and quantitatively, that these structural limitations indeed constrain peripheral healthcare providers in their provision of MCH services, to such an extent that an even marginal improvement of their working conditions through regular supervisory visits provided during the study, was associated with a general improvement of the follow-up of HIV-exposed infants. Any potential effect of the reorganisation, integrating MCH service delivery, could not be shown in our study in the face of such limitations. We recommend therefore that the improvement of the basic working conditions of peripheral MCH staff in public healthcare facilities in Mozambique and similar settings, including regular support and an adequate supply of commodities, might be a more effective way to improve the follow-up of HIV-exposed infants, before considering any reorganization of service delivery.