CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 4, May 2013
128
AFRICA
Discussion
A patient with rheumatic heart disease is expected to receive
at least 80% of the annual prescribed injections. Receiving less
than 80% of the injections places an individual at a higher risk
of recurrent ARF and its complications.
9
In this study, adherence
was considered as when a patient had received at least 80% of the
required injections over a period of six months.
An adherence level of at least 80% was found among 44
(54%) patients, compared to 38 (46%) with adherence levels
less than 80%. The mean adherence was determined at 70.12%
(SD 29.25). This was similar to the adherence level determined
by Harrington in an aboriginal community in Australia, in which
59% of patients had received more than 75% of their prescribed
injections during an interview.
14
However, the level of adherence we determined in this study
was considerably higher than that found among RHD patients
in another Aboriginal community in Australia were the mean
adherence level was 56% when patients were followed up for a
period of 24 months.
15
On the other hand, this level of adherence
was considerably less than that found in several other studies
such as the study done in Haryana district in India, which
found that 90% of the patients had received over 80% of their
benzathine injections over the previous eight years.
16
The variability in levels of adherence may reflect the different
systems in which these studies were done, duration of follow up,
the different factors that may influence adherence, the individual
study designs, and the different cut-off points for defining
adherence in the different studies. This variability is still hard
to explain confidently since low levels of adherence have been
demonstrated in Australia were rheumatic heart disease registries
exist and are fully functional. However, given all these factors,
the level of adherence as was determined in this study was low,
because it placed a significant proportion of patients (46%) at
risk of recurrent episodes of ARF and worsening of valvular
heart lesions, with resultant poor prognosis.
Analysing the association of patient factors with adherence
levels provides insight into those groups at particular risk of
recurrence through poor adherence. No particular patient factor
was found to be significantly associated with adherence. This was
not surprising as a similar study by Stewart in Australia found no
significant demographic factors associated with adherence.
15
However, trends towards adherence were demonstrated among
patients who resided in a town/city and those with at least
secondary level of education.
The fact that patients who resided in a town/city tended to
have better adherence could be explained by the fact that these
patients have easier access to healthcare facilities compared
to those from rural areas. This finding could be supported by
a study done by Kathie Walker, who found that patients who
stayed far from the health facilities (
>
10 km) were significantly
associated with poor compliance.
11
There was no difference
between men and women regarding their level of adherence,
although an earlier study by Dorothy
12
had revealed that men
are more likely to be non-adherent compared to their female
counterparts.
Whether the lack of significant factors reflects a true lack of
association, a limited time to follow up, or rather, the effect of
a small sample size is uncertain from these results. Most of the
studies done on this topic have not analysed for these patient
factors, making it rather a complex area to discuss. Nevertheless,
these trends do identify subgroups that might be at increased risk
of recurrent ARF and worsening of RHD through non-adherence.
The commonest reason reported for missing monthly
benzathine prophylaxis injections was the painful nature of
the injection (27, 29%). This was closely followed by lack of
money (26.9%) and the fact that the patients felt healthy and
well (11.8%). These factors have also been described by WHO
expert consultation in Geneva.
6
Despite some reports which
have indicated that forgetting could be an import reason for
missing injections, it did not feature in this study. This could
have been due to the benzathine penicillin card but it cannot be
ascertained for sure, since no control group existed. However,
other factors of interest included development of injection
abscesses, misperceptions by the local health worker that a
patient does not need prophylaxis after heart valve surgery, and
missing an injection during admission. These factors will form
the basis for intervention in order to improve adherence among
our patients.
Limitations
Providing a card, which is not routinely done in the normal
setting, may have improved adherence rates as this could have
acted as a reminder to go for the injection although the card was
devoid of reminding dates. In our setting where records are poor,
we could not find any other objective way to measure these rates.
The results obtained may not apply to the general population
because of the sampling procedure used. The follow-up time
of six months may have been too short to accurately assess the
levels of adherence since adherence has been shown to decline
with time in some studies.
Conclusions
Although the mean level of adherence was fairly good, the level
of non-adherence among these rheumatic heart disease patients
was significant. Although no particular patient factor was found
to be significantly associated with adherence, we determined
that residing in a town/city and having at least a secondary
school level of education was associated with a trend towards
adherence. The painful nature of the benzathine penicillin
injections and lack of transport money to travel to the health
centre were the main reasons for non-adherence among RHD
patients attending Mulago Hospital.
This study was supported by a postgraduate research grant from the Uganda
National Council for Science and Technology under the Millennium Science
Initiative and the Uganda Heart Institute. The expert technical assistance of
Elias Sebatta is gratefully acknowledged. We thank Ms Beatrice Mwesige
and Gladys Kahima of the Echocardiography laboratory and the entire staff
and management of the Uganda Heart Institute for their support.
References
1.
Consenso s. Prevenção da Febre Reumática da Sociedade Brasileira de
Pediatria.
Correios
da
SBP
2004;
10
: 9–14.
2.
Rayamajhi A, Sharma D, Shakya U. First-episode versus recurrent acute
rheumatic fever: is it different?
Pediatr
Int
2009;
51
(2): 269–275.
3.
Carapetis JR, McDonald M, Wilson NJ. Acute rheumatic fever.
Lancet
2005;
366
(9480): 155–168.
4.
Steer AC, Carapetis JR. Acute rheumatic fever and rheumatic heart
disease in indigenous populations.
Pediatr Clin North Am
2009;
56
(6):
