Selection criteria and search

Our search for delayed leprosy case detection was based on (a) the period of delay calculated from the beginning of signs or symptoms to diagnosis, either in numerical or categorical values; and (b) the occurrence of G2D. To identify factors determining delayed detection, we used PICO (Population, Intervention/Exposure, Comparison and Outcome) to make structured inclusion and exclusion criteria. Population: leprosy patients in leprosy high and low endemic countries were included. Exposure: Health care services factors. Comparison: we included all studies which presented case detection delay as expressed in time periods of months and as cases categorized as G2D without comparing the health care services factors that contributed to them. Outcome: the disability in leprosy. We systematically searched five electronic databases: Embase, Medline All Ovid, Web of Science, Cochrane CENTRAL, and the WHO Global Health Library (for details of the search strategy, see S1 Text). We included original empirical leprosy-related studies published in English between January 1, 2000, and January 31, 2021. We excluded case reports, articles without full text (abstract only), and articles that made no reference either to delayed case detection or to factors associated with it.

Data extraction

To select articles for full-text screening, two reviewers (YD and AF) independently screened article titles and abstracts. Data from articles were extracted and double-entered into Microsoft Excel. Disagreements were settled by a third reviewer (IK or JHR). The extracted data included author(s), year of publication, article title, journal title, study design, study setting, number of study participants, and summaries of the relationship between healthcare factors and delays in leprosy case detection.

Data analysis

After reviewing all the possible healthcare factors, we listed them as variables associated with delays in leprosy case detection. These variables were then categorized into three main groups: 1) Structural factors; 2) Health service factors; and 3) Intermediate factors by considering causal pathways written in each study reviewed. In the quantitative studies, the strength of the association between factors was expressed as Odds Ratios (ORs), adjusted Odds Ratios (aORs), Hazard Ratios (HRs), and/or significance p-values. In the qualitative studies, we summarized the content of the studies that stated a relationship between healthcare factors and delayed leprosy case-detection or G2D. We conducted a narrative analysis. A meta-analysis was not feasible due to the variation in the methods as applied in the included studies. The methods and results have been reported following the PRISMA guidelines (see S1 Table).

Evaluation of the quality of studies

The quality of articles was assessed using a risk-of-bias instrument for potential study-design biases. For quantitative studies, a scoring checklist was used to assess the quality of the research hypothesis, the study population, selection bias, exposure, outcome, and confounding; and to formulate an overall opinion of the study’s validity and applicability through the accumulation score of the assessment of the method, sampling, data, result, ethics, implications, and usefulness of the study. This assessment was used for all papers that have been reviewed, which all have observational study design [15]. For qualitative studies, a COREQ checklist was used to evaluate the research team and reflexivity; the study design; and our analysis and findings [16]. For mixed-method studies, both methods were combined. Quality was evaluated by two reviewers (YD and AF). In cases of disagreement, a third reviewer (IK or JHR) was invited to resolve the issue.

1. Structural factors

Structural factors are basic health-service-related factors that contribute to delays in case detection. They include i) financial and logistic issues, and geographical circumstance (which we classified as barriers); ii) health service organization and management including of the level of decentralization of healthcare services (classified as facilitators).

Two studies in Peru and Brazil [30,35] both published in 2020 reported delays related to poor geographical access to healthcare services. Another barrier described by one study post-elimination in Myanmar (2020) was related to lack of financial and logistic support, which limited community-based leprosy field activities [36]; health staff faced high transportation costs for supervising the leprosy control program, and relied on other health programs for field activities [36]. One study in Peru (2010) that evaluated pre and post the leprosy control program decentralization reported that the decentralization of leprosy control and its integration into primary healthcare was a facilitator that led to significantly earlier case detection (p = 0.012) [20]. Decentralization led people with leprosy to visit health services closer to their home with percentages increasing from 23·5% to 46·6% (p<0.001), which led in turn to an increase in self-reporting from 38·7% to 51·1% (p = 0.010) and to a reduction of G2D leprosy at diagnosis from 16.2% to 7.6% (p = 0.012). In contrast, disruptions of decentralized leprosy care, such as a high turnover of health staff and low commitment, contributed to longer case-detection intervals [35], as reported in one study in Brasil (2020) that aimed to analyze assistance provided to people affected leprosy by healthcare management. One paper reported [32] that, in Nepal, the implementation of fully integrated leprosy services contributed to earlier case detection than the earlier non-integrated services, when people visited specialist services as only option for care, often leading to long delays. The low endemicity of leprosy can also contribute to delays in detection: one qualitative study with mix method from India that was done in three districts in one state [31] reported increased detection delays and a lack of involvement by health staff in the National Leprosy Elimination Program (NLEP) after leprosy was declared to have been “eliminated” or “controlled” by the central government.

Fig 2 includes a further structural factor: health-service organization and management. Although this umbrella term was not mentioned explicitly in the articles reviewed, there are many allusions to it.

2. Health services factors

Our review found many Health service factors related to healthcare and human resources that contributed to delayed case detection (Fig 2). One study reported that the limited number of referral centers for providing appropriate leprosy care [36] was related to such delays. Regarding personnel resources, three studies [30,35,36] reported that the limited availability of healthcare professionals contributed to delays. One study [30] reported that a lack of specialists caused by geographical barriers was associated with delays, as a result of which people with leprosy perceived health services for leprosy to be limited and geographical access to them difficult. One paper [36] stated that, after the completion of leprosy-elimination programs, many members of health staff mentioned that delays had been related to a lack not only to of local leadership, but also of interest in and commitment to leprosy. Another qualitative study [35] mentioned that high personnel turnover, lack of physical resources, and healthcare professionals’ low commitment could contribute to detection delays. The same study also reported that prejudice about leprosy among basic healthcare professionals often led to unnecessary referrals and thereby delays [35].

Regarding community perceptions and activities in the leprosy-care services, several studies reported leprosy patients’ perceptions of health workers’ knowledge and skills and the quality of health services [22,29–31,35]. One qualitative study [31] reported on the important role played by community volunteers in the public health sector, whose involvement helped reduce delays through the early identification and referral of patients to a health facility.

Two studies reported that lack of knowledge regarding leprosy reactions (a complication of leprosy) were an indication of health staff’s unfamiliarity with leprosy [22,34]; as staff had difficulty diagnosing leprosy, they contributed to detection delays [22]. Two studies investigating health staff’s diagnostic capabilities revealed patients’ perceptions that health staff did not know what leprosy was [29], and described health professionals at different levels of the healthcare system as “remarkably unprepared to diagnose leprosy”[35]. Two qualitative studies reported patients’ feelings about the cost of health services, their fear of the diagnosis, the uncertainty of treatment [29], and the mistrust they encountered at the public health facilities [31]. Speaking about the health services, patients referred to a “lack of compassion among doctors” and “long waiting [times]”; and said “They don’t treat us good in the government hospital”, “I don’t like to go to Public Health Centers”, and “doctors should be kind to patients” [31].

3. Intermediate factors

We identified three intermediate factors that contributed to delayed case detection (Fig 2). These factors occur in a causal pathway from health service factors to case detection delay. The first was misdiagnosis leading to mistreatment. Misdiagnosis was often cited as a factor in detection delays [17,21,22,26,29,30,32,36]. The misdiagnosis that was reported most was arthritis, photosensitivity, nerve disease or other skin diseases [32]. One study [26] reported that for periods lasting between three months and 10 years before being diagnosed with leprosy, 32.5% of respondents had been living with a diagnosis other than leprosy. Two studies reported that detection had been delayed in 70% [17] and 65% [32] of their patients.

One study [19] reported that patients needing treatment for leprosy reactions (a severe complication of leprosy that often leads to nerve damage) had had shorter delays at the first intake than those without leprosy reactions (p<0.01). This was because leprosy reactions often cause severe acute symptoms such as pain, swelling of patches, and fever, which drove patients to seek medical attention quickly.

Misdiagnosis can occur because of non-specific symptoms of leprosy, or even of being asymptomatic in the initial stage of disease, and because of healthcare staff’s low awareness of the initial appearance of leprosy. A study from the UK [25] reported that 6 out of 12 leprosy cases came with unusual presentations, such as pure neuritic leprosy, trigeminal nerve involvement, and anesthetic patches on the thigh, and that these led to misdiagnosis. Another study from Brazil [17] reported that 4.8% of leprosy cases were asymptomatic and diagnosed accidentally.

The second intermediate factor in detection delays was the number of consultations and examinations–physical and otherwise–that were needed before leprosy was diagnosed. Six studies investigated the relationship between delayed case detection and the number of examinations and/or consultations in healthcare services [17,18,21,23,26,33]. On average, the detection delay for patients who needed five or more consultations was 24.4 months longer than for those diagnosed during the initial consultation (p = 0.009) [17]. Repeated contact with health-service practitioners was associated with delayed detection (OR: 2.05, 95% CI: 1.42–2.97, p<0.001) [18]. For every additional examination, the odds of a patient that would receive a quicker diagnosis of leprosy were 46% (OR: 0.539, 95% CI: 0.393–0,739, p<0.001) [21]. One study reported that many patients (45.5%) had been diagnosed at their second to fifth visit [23]. Another study reported that 30% of patients in its sample of cases had needed more than five visits to medical services before the diagnosis was suspected [26]. The delay for those with 1–3 visits was less than it was for those with four or more visits (p = 0.051) [33]. In contrast, two studies [17,21] reported that examinations confirming the diagnosis were not associated with detection delay.

The third intermediate factor in detection delays related to the type of healthcare service. Lack of referral services or the existence of integrated leprosy services could lead patients to visit nearby health services that had only limited experience of diagnosing leprosy. Three studies [18,28,31] reported statistically significant relationships between delayed case detection and the type of healthcare service patients had visited. Delayed case detection was associated with a first consultation at a general health service (aOR:2.10, 95% CI: 1.46–3.00, p<0.001) [18]; at another service than a skin clinic (OR: 3.21, 95% CI: 1.68–6.14, p< 0.001) [28]; and at a private healthcare provider (aOR:2.6, 95% CI: 1.4–5.2, p<0.05) [31]. One study [21] reported that health system delays were not significantly associated with referral to another doctor (OR: 1.667, 95% CI: 0.8–3.4, p = 0.176).