Task shifting in the management of hypertension in Kinshasa, Democratic Republic of Congo: a cross-sectional study

We conducted a cross-sectional study in Kinshasa Primary Health-Care (KPHC) network facilities from December 2015 to January 2016.The KPHC network consists of 51 facilities (11 GRHs and 40 HCs) that serve patients living in Kinshasa [21].

The study included patients with an HTN diagnosis made by a healthcare provider and who were receiving care in the KPHC network facilities. Patients were eligible for the study if they were aged 18 years and older, presented for care in a participating health facilities on the day of the survey, and agreed to participate in the study.

The sample size was computed using the following formula:

Where p1 represents the proportion of patients with controlled HTN at the HC level (15.6%) [19]; p2 represents the proportion of patients with controlled HTN in the GRH level (we assumed that the proportion would be twice that in HCs (31.2%); z is the value of the standard normal distribution corresponding to a significance level of alpha of 0.05 (1.96); and β is power (80%), z1-β = 0.84.

The minimal sample size computed was 112 patients per group or 224 patients for both groups. To recruit patients, we obtained the list of GRHs and HCs from the KPHC network and visited all them to know the dates when HTN patients have follow-up clinics. We then visited the facilities on clinic dates and successively recruited patients until the desired sample size was achieved. The minimum sample size was achieved after visiting nine HCs and five GRHs.

Five trained data collectors’ conducted face-to-face interviews using a structured questionnaire. The questionnaire was pretested and translated into the local language (Lingala) before data collection. The questionnaire elicited information on the following variables: patients’ socio-demographic (sex, age, marital status, educational level, income) and clinical characteristics (duration of HTN, co-morbidity); health system variables (patient-provider relationship, the time devoted to consultation, waiting time, affordability and availability of healthcare); treatment related variables (types of antihypertensive drugs and dosage, experience of medication side effects, treatment adherence) and medical-related costs (consultation fees, medicines, laboratory, transportation and food).

The dependent variable was uncontrolled HTN, which was defined as having a systolic blood pressure (SBP) ≥ 140 mmHg and/or a diastolic blood pressure (DBP) ≥90 mmHg for patients without co-morbidity or having SBP ≥ 130 mmHg and/or DBP ≥ 80 mmHg for patients with co-morbidity [22]. The primary explanatory variable, task shifting, was defined as the management of HTN at HCs level.

Co-morbidity was defined as HTN associated with diabetes mellitus or target organ damage (heart disease, stroke, or chronic kidney disease) based on diagnosis made by a healthcare provider. Antihypertensive types were categorized in two groups, monotherapy if a patient was on one type of antihypertensive medication and multitherapy if a patient was on two or more types of antihypertensive medications. Treatment adherence was measured using the Morisky Scale, which is a validated four-item scale with good internal consistency based on studies of inner-city patients with HTN [23, 24]. Patients responded “yes” or “no” to four questions (Do you ever forget to take your medicine? Are you careless at times about taking your medicine? When you feel better, do you sometimes stop taking your medicine? Sometimes if you feel worse when you take the medicine, do you stop taking it?). Based on their responses, patients were categorized into three groups: high, medium and low adherence. To facilitate statistical analysis, adherence was dichotomized with patients with low or medium adherence classified as non-adherent and patients with high adherence classified as adherent [25].

The costs of medicines were reported on a monthly basis. The transportation costs comprised roundtrip travel costs incurred by the patient and anyone who accompanied them to the health facility for HTN-related care. Food costs comprised expenditures related to food during HTN-related health facility visits. Consultation, medicines and laboratory costs were obtained from patient record notebooks, invoices and bills provided by patients. All KPHC patients have a patient record notebook where all information related to their care including costs are recorded. Transportation and food related costs were self-reported. The cost computations did not include productivity and opportunity costs. Costs were provided in Congolese Democratic Francs (CDF), the local currency, and converted into United States Dollars (US$) using the average exchange rate during the study period (920 CDF to US$ 1). To account for inflation, a rate of 2% was used in computing all costs incurred outside the reference period (2015 to 2017).

Data were entered into Epi data then exported to SPSS (Statistical Package for Social Sciences) version 20.0 (SPSS, Inc., Chicago, IL, USA) and Microsoft Excel for analysis. Descriptive statistics were used to summarize patients’ characteristics. Categorical variables were reported as frequencies and percentages. Continuous variables were reported using means with standard deviation. The Student’s t test and χ² test were used to compare means and proportions respectively. Logistic regression was used to examine the association between task shifting and uncontrolled HTN after adjusting for others factors like socio-demographic, clinical factors. For all analyses, a p-value of less than 0.05 was considered statistically significant.

The study protocol was reviewed and approved by the institutional review board of the Kinshasa School of Public Health. All study participants provided written informed consent.

Patients’ socio-demographic and clinical characteristics are summarized in Table 1. Almost half of the patients were female (53.1%). The patients’ mean age was 59.5 years (standard deviation = 11.4 years). A significantly higher proportion of GRH patients than HC patients had ever attended school (87.7% vs 77.7%, p = 0.035). Co-morbidity was significantly more common in GRH patients compared with HC patients (45.4% vs 25.4%, p < 0.001). The majority (94.2%) of patients had been diagnosed with HTN within the 5 years preceding the study. Only 23% of all patients had controlled HTN. The proportion of patients with uncontrolled HTN did not differ by type of health facility (76.2% among GRH patients vs 77.7% among HC patients, p = 0.771).

Patients’ assessments of the quality of care are summarized in Table 2. A significantly greater proportion of GRH patients than HC patients stated that the consultation time was sufficient (96.9% vs 81.5%, p < 0.001). No statistically significant differences between GRH and HC patients were observed in terms of perceived accessibility, waiting times, affordability, satisfaction with the relationship with healthcare providers, types of antihypertensive drugs used, adherence to medication, and control of HTN. A greater proportion of GRH patients than HC patients reported medication side effects (22.6% vs 8.3%, p = 0.003). The proportion of patients who were not adherent to medication did not differ by facility (GRH 47.2% versus HC 38.5%, p = 0.204). No significant difference was also found between the two groups with regard to the control of HTN.

Results of the logistic regression model employed to assess the factors associated with HTN control are summarized in Table 3. The type of facility was not significantly associated with HTN control. Co-morbidity and the type of antihypertensive medication used were associated with HTN control. Patients with co-morbidity were more likely than those without co-morbidity to have uncontrolled HTN (adjusted OR = 10.3; 95% CI: 3.8–28.3) while patients on multiple medications were more likely than those on a single medication to have uncontrolled HTN (adjusted OR = 4.6; 95% CI: 1.3–16.1).

Table 4 shows the results of the cost analysis. In the HC, the most expensive costs were medication costs (US$4.4 [95% CI: US$3.4 – US$5.4]) while in the GRH the most expensive costs were laboratory costs (US$23.6 [95% CI: US$ 16.9 – US$ 30.3]).The mean total cost was lower in the HC than in the GRH (US$ 7.7 [95% CI: US$ 6.6 – US$ 8.8] versus US$34.2 [95% CI: US$27.6 – US$40.9]).