The journey of cancer patients and the quest to equity: findings from Morocco

Abstract

Objectives

Rapid diagnostic and assessment pathways for cancer patients provide timely and effective care. This study took place in Morocco, where the majority of patients treated in the public sector are diagnosed at an advanced stage. The aim of this study was to determine the duration of different time intervals along the cancer patient pathway and to highlight problem areas so that strategies can be implemented to make the process more equitable and effective.

Study design

Cross-sectional study.

Methods

Recently diagnosed cancer patients were recruited from four major oncology centres in Morocco; namely, Marrakech, Casablanca, Rabat, and Fez. A questionnaire survey was administered, including sociodemographic and medical information and questions on access to the oncology centre, beliefs, and opinions on the medical staff. The dates of symptom recognition, assessment, diagnosis referral, biopsy, and treatment initiation were collected. Different time intervals (patient, diagnosis, biopsy, and treatment) were estimated and their determinants were investigated.

Results

A total of 812 patients were interviewed. The majority of participants were breast cancer patients. In total, 60% of participants were at stage III–IV. The main facilitators of cancer diagnosis confirmation and treatment initiation were easy access to diagnosis and treatment facilities, financial resources, personal history of cancer, time availability, late stage at diagnosis, advanced age, and private health insurance. The patient interval (i.e., time from symptom recognition to initial healthcare assessment) had a median duration of 30 days. The biopsy and treatment intervals were within the current international recommendations (7 and 28 days, respectively). However, the diagnosis interval (52 days) was twice as long as the recommended timeframes from the UK, Australia, and the World Health Organization (<28 days).

Conclusions

Interval targets should be defined to encourage health systems to be more equitable and effective and to ensure that cancer patients are treated within a defined timeframe.

Introduction

The journey of a cancer patient through the health system is marked by a series of interactions leading to cancer diagnosis and treatment. Depending on the setting, various obstacles may exist at different stages, from seeking initial healthcare advice, to receiving a diagnosis (after a series of referrals to specialists, imaging and biological tests), to eventually attending an oncology centre to start treatment. The four major steps of this journey (i.e., ‘appraisal’, ‘help-seeking’, ‘diagnostic’, and ‘treatment’)¹ enable the healthcare pathway to be defined into specific intervals. Duration of these intervals vary according to patient characteristics, disease features, and performances of both the providers and the health system.

Several systematic reviews have reported on the duration of diagnosis and treatment intervals and on their determinants.^{1, 2, 3} Another comprehensive systematic review confirmed that short diagnosis and treatment intervals had beneficial clinical outcomes in breast and colorectal cancer patients.⁴ Indeed, it seems fair to presume that efforts to speed up the diagnosis of symptomatic patients are likely to have benefits in terms of earlier stage at diagnosis, improved survival, and improved quality of life.⁴ Rapid diagnostic and assessment pathways illustrate how timely and effective care can be provided to patients presenting with cancer symptoms.^5,6 Moreover, several countries put interval targets in place to encourage the health system to be more equitable⁷; however, these targets are mainly related to diagnosis and treatment intervals only.

The Moroccan Ministry of Health, in close collaboration with the Lalla Salma Foundation, has put effort in to the different aspects of cancer control.^8,9 Despite these efforts, the majority of patients hospitalised in public oncology centres in Morocco present with an advanced stage at diagnosis. This study investigated the pathway from symptom recognition to treatment initiation, including the duration of the main intervals and the determinants of these intervals. The interval duration results from this study, when compared with international standards, provide evidence on how to improve and speed up the referral for treatment of cancer patients in Morocco.

Methods

This multicentre, cross-sectional study used a face-to-face questionnaire. Study participants were cancer patients being hospitalised for treatment in one of the four main oncology centres in Morocco (Marrakech, Casablanca, Rabat and Fez).

Participant inclusion criteria were as follows: (1) adult patients with histologically confirmed cancer of any anatomical site; (2) able to answer questions in Arabic or French; (3) provided informed consent. Paediatric cancer cases were excluded (aged 0–14 years).

The interviewers and the principal investigator (PI) of each oncology centre were trained at the same time during a 2-day session by the Casablanca PI (KB). The objective of the training was to standardise the administration of the questionnaire and to define the information to be collected for each item, especially the dates. Data entry via the study portal was also demonstrated.

All newly registered patients were approached and invited to take part in the research project. Consenting study participants were recruited during a 1-month period. The Casablanca centre recruited participants from 1 to 28 February 2018, and the Marrakech, Rabat and Fez centres recruited eligible patients from 1 to 31 May 2018.

Interviews took place inside the oncology centre and lasted 20–30 min. The questionnaire included items on patient identification; sociodemographic information (education level, occupation, marital status, number of children, usual place of residence, place of living during the cancer treatment); medical information (personal history of cancer, direct family history of cancer [parents/children/siblings], cancer stage, cancer site, symptoms duration before visiting a primary health facility); duration to reach a primary health centre, a diagnostic centre, and the oncology centre; information on access to the oncology centre (transportation means, opinion on access using the 5-point Likert scale [1 = strongly agree, 2 = agree, 3 = do not agree nor disagree; 4 = disagree, 5 = strongly disagree]); financial information (health insurance, monthly income, family dependence status, household belongings, global cost of cancer, opinion on the economical aspect of the cancer management); beliefs; and opinions on the attitude of the medical staff before referral to the oncology centre. Study data were collected and were managed using Research Electronic Data Capture hosted at the International Agency for Research on Cancer (IARC).^10,11

The following key dates were requested in order to allow interval duration estimates: date of symptom recognition, date of biopsy prescription, date of biopsy, date of biopsy result, and date of admission to the oncology centre (representing the date of treatment initiation). The date of symptom recognition was based on information from the referral letter to the oncology centre, or it was recollected in relation to familial, secular, or religious celebrations (e.g., the Ramadan or the Eid Festival). Other dates were collected from the oncology centre patient file or from the electronic medical record.

The main analyses included the determinants of four major intervals. The ‘patient interval’ was defined as the duration between symptom recognition and first presentation to a healthcare professional to seek evaluation. Symptom duration was requested only at the Casablanca centre. Since the date of symptom recognition was available, the date of the first presentation to a medical provider was estimated by adding the symptom duration date to the date of symptom recognition. Indeed, this date of first presentation could only be extrapolated because patients rarely remember the exact date when they visited a primary healthcare centre to discuss the related symptoms. The ‘diagnosis interval’ was defined as the duration between the estimated date of first consultation and the date of cancer diagnosis (i.e., date of the histology report with a cancer diagnosis). This diagnosis interval was available only from the Casablanca centre. The ‘treatment interval’ was defined as the time between the cancer diagnosis and hospitalisation for treatment at the oncology centre. This interval was estimated for the four oncology centres.

Median durations of other intervals were also computed. They included median duration between biopsy prescription and biopsy sampling, median time between biopsy sampling and histopathology result, and median interval between first consultation and hospitalisation for treatment (or ‘health system interval’). Durations are presented as medians with interquartile ranges (IQRs). All intervals can be found in Fig. 1.

Fig. 1.

Median duration and interquartile range (in days) of the different intervals of cancer patients' journey.

Statistical analyses

Patient sociodemographic characteristics are presented as proportions. Median duration in days was estimated for the different intervals. The effects of potential determinants on the patient interval, diagnosis interval and treatment interval were estimated by obtaining relative risks (RRs) and their 95% credible intervals (CIs) using Bayesian exponential regression models. Multivariate analysis was performed by including the patient sociodemographic characteristics in the same regression model. Due to the large number of variables on the opinions about oncology hospital access, economical aspects of cancer, beliefs, and attitudes of medical staff, the results would be difficult to interpret; thus, principle component analysis (PCA) was used to create interrelationships and provide fewer continuous variables (components) that were assigned meaningful descriptive statements and stored as component scores.

For easy interpretation of the component scores, the Likert scale of the variables on opinion included in the PCA was recategorised as follows: 1 = strongly disagree; 2 = disagree; 3 = undecided; 4 = agree; and 5 = strongly agree. High-positive patient-component scores signified patient agreement with the described component and high-negative scores signified disagreement. To avoid the problem of collinearity when many highly correlated independent variables are in the regression model, the few uncorrelated computed components were included, and these were additionally adjusted for in the multivariate regression models. Significant differences were inferred at a 5% level of significance. Statistical analyses were carried out in Stata 17.0 (Stata Corp LP, Texas, USA) and the Just Another Gibbs Sampler software.

Results

A total of 812 patients were recruited for this study, mainly from Casablanca (n = 234; 29%) and Rabat (n = 229; 28%), followed by Fez (n = 195; 24%) and Marrakech (n = 154; 19%). The majority of patients were women (66%), and the main cancer sites were breast (29%), cervical and colorectal (9% each), and lung (8%). Among the 672 patients with stage information, 60% were at Stage III or IV (see Supplement Table S1).

Fig. 1 shows the median duration and interquartile range of the different intervals. The median durations were as follows: 30 days to first seek medical advice after symptom recognition; 32 days for healthcare providers to prescribe a biopsy; 3 days to have a biopsy appointment; and 7 days to receive the histopathology results. The overall diagnosis interval was 52 days from the date of the first presentation to a healthcare provider, and the median duration between symptom recognition and cancer diagnosis was 141 days. The treatment interval (i.e., time between histological confirmation of cancer and treatment initiation at the oncology centre) was 28 days (range: 12–65 days), and the health system interval (representing the duration from the first consultation to initiate treatment after the mandatory histological confirmation) was 90 days. Median durations of each interval according to major determinants can be found in the Supplement Table S2.

Table 1 presents the determinants of the patient interval duration. A shorter patient interval was statistically significantly associated with older age, being currently married, being in the highest household income category, living in a suburban area, and being covered by private or civil servant health insurance. Having five or more children, compared to none or one child, increased the patient interval duration by 3-fold. Similarly, individuals with a personal history of cancer had a patient interval that was twice as long as individuals without a cancer history. This study did not find any differences in patient interval duration related to gender or education. Belief that “if an abnormality is not disturbing, there is no need to consult a doctor” or “the symptoms are benign and will clear” were not associated with patient interval (adjusted RR for duration as continuous variable: 1.01 [95% CI: 0.92–1.09]).

Table 1.

Determinants of patient interval^a (Casablanca centre data only).

Characteristics	Patients assessed (n = 230)	Crude relative risk (95% CI)		Adjusted relative risk (95% CI)
Gender
Men	68	1.00	(Reference)	1.00	(Reference)
Women	162	0.81	(0.60–1.06)	0.75	(0.47–1.08)
Age group (yrs)
16–49	76	1.00	(Reference)	1.00	(Reference)
50–69	113	0.84	(0.61–1.10)	0.67	(0.44–0.94)
70+	38	0.67	(0.43–0.97)	0.40	(0.20–0.69)
Education
No	141	1.00	(Reference)	1.00	(Reference)
Yes (primary, secondary, college)	89	1.24	(0.93–1.59)	1.43	(0.95–1.99)
Currently married
No	83	1.00	(Reference)	1.00	(Reference)
Yes	141	1.08	(0.79–1.38)	0.71	(0.47–1.01)
Total no. of children
0–1	50	1.00	(Reference)	1.00	(Reference)
2–4	92	0.88	(0.57–1.22)	1.24	(0.72–1.87)
≥5	75	1.29	(0.85–1.83)	3.09	(1.54–5.15)
Currently employed
No	196	1.00	(Reference)	1.00	(Reference)
Yes	33	0.80	(0.53–1.14)	0.92	(0.50–1.46)
Monthly household income (MAD)
None	75	1.00	(Reference)	1.00	(Reference)
<2000	107	1.46	(1.06–1.90)	1.48	(1.00–2.12)
2000–4000	36	1.48	(0.95–2.13)	1.14	(0.64–1.80)
>4000	8	0.39	(0.16–0.77)	0.38	(0.12–0.83)
Place of residence
Urban	139	1.00	(Reference)	1.00	(Reference)
Suburban	41	0.67	(0.46–0.93)	0.43	(0.24–0.68)
Rural	49	1.52	(1.06–2.06)	0.91	(0.53–1.43)
Type of health insurance
Basic health insurance (RAMED)	191	1.00	(Reference)	1.00	(Reference)
Civil servants and private insurance	27	0.58	(0.37–0.84)	0.52	(0.31–0.82)
None	8	1.79	(0.74–3.45)	2.33	(0.83–4.97)
Personal history of cancer
No	212	1.00	(Reference)	1.00	(Reference)
Yes	14	2.63	(1.44–4.41)	2.53	(1.14–4.49)
Direct family history of cancer
No	197	1.00	(Reference)	1.00	(Reference)
Yes	31	0.66	(0.43–0.94)	0.57	(0.36–0.85)
Belief that the disease can be cured by a healer or maraboutb	223	1.16	(1.01–1.31)	1.09	(0.90–1.31)
Anomaly or symptoms are minor, hence no need for medical consultationb	223	1.02	(0.95–1.09)	1.01	(0.92–1.09)

CI: confidence interval; MAD, Moroccan dirham.

^aPatient interval is defined as the time from symptom recognition to presentation at clinic/health professional.

^bComponent scores created using principle component analysis with low scores indicating strongly agree and high scores strongly disagree.

The determinants of diagnosis interval duration are presented in Table 2. Compared to men, women had a 3-fold longer duration to receive a cancer diagnosis. Patients covered by private or civil servant health insurance also had a longer diagnosis time than deprived and vulnerable patients covered with the RAMED (or basic health insurance). On the other hand, being aged ≥70 years, having been to school (compared to no education), having a caregiver to coordinate the diagnosis period (i.e., a navigator), and being in the highest household income categories (compared to households with no income) reduced the diagnosis interval. Patients who felt that they had been blocked from getting an imaging appointment, compared to those who did not have this feeling, were associated with a 2-fold longer diagnosis interval. However, no significant association was observed with the feeling of being blocked for appointments with doctors or for a biopsy. Patients who generally agreed with the statements that “if an abnormality is not disturbing, there is no need to consult a doctor” or “the symptoms are benign and will clear” had a shorter diagnosis interval (RR = 0.81; 95% CI = 0.73–0.90). However, the more they agreed that “cancer is a punishment”, “cancer patients will be rejected by people around them”, or “cancer can be treated by healers or marabouts”, the longer was the diagnosis interval (RR = 1.43; 95% CI = 1.20–1.70). Factors related to the financial impact (i.e., opinion on the cost of exams, transportation, need of financial support from donors or relatives), the health providers behaviour (i.e., not enough explanation, delayed referral, malpractice), and access to healthcare facilities were not significantly associated with the duration to receive a cancer diagnosis.

Table 2.

Determinants of diagnosis interval^a (Casablanca centre data only).

Characteristics	Patients assessed (n = 215)	Crude relative risk (95% CI)		Adjusted relative risk (95% CI)
Gender
Men	59	1.00	(Reference)	1.00	(Reference)
Women	156	1.42	(1.02–1.88)	2.84	(1.65–4.25)
Age group (yrs)
16–49	71	1.00	(Reference)	1.00	(Reference)
50–69	106	1.96	(1.42–2.60)	1.16	(0.66–1.79)
70+	35	1.36	(0.88–1.99)	0.53	(0.23–0.99)
Education
No	134	1.00	(Reference)	1.00	(Reference)
Yes (primary, secondary, college)	81	0.43	(0.32–0.55)	0.48	(0.28–0.74)
Currently married
Not	78	1.00	(Reference)	1.00	(Reference)
Yes	131	0.71	(0.53–0.92)	1.31	(0.71–2.09)
Total no. of children
0–1	47	1.00	(Reference)	1.00	(Reference)
2–4	83	0.98	(0.64–1.36)	0.56	(0.25–1.05)
≥ 5	72	1.55	(1.04–2.19)	0.89	(0.32–1.85)
Currently employed
No	184	1.00	(Reference)	1.00	(Reference)
Yes	30	1.53	(0.99–2.24)	1.63	(0.93–2.62)
Monthly household income (MAD)
None	72	1.00	(Reference)	1.00	(Reference)
<2000	101	1.39	(1.01–1.86)	0.72	(0.43–1.11)
2000–4000	34	0.48	(0.31–0.71)	0.22	(0.10–0.40)
> 4000	6	0.32	(0.12–0.74)	0.29	(0.05–0.82)
Place of residence
Urban	126	1.00	(Reference)	1.00	(Reference)
Suburban	41	0.74	(0.50–1.04)	0.82	(0.47–1.25)
Rural	47	1.11	(0.76–1.54)	0.76	(0.38–1.27)
Type of health insurance
Basic health insurance (RAMED)	183	1.00	(Reference)	1.00	(Reference)
Civil servants and private insurance	21	1.40	(0.83–2.16)	2.21	(1.08–3.86)
None	8	0.89	(0.37–1.75)	2.52	(0.91–5.48)
Personal history of cancer
No	199	1.00	(Reference)	1.00	(Reference)
Yes	13	0.10	(0.05–0.17)	0.06	(0.02–0.13)
Direct family history of cancer
No	184	1.00	(Reference)	1.00	(Reference)
Yes	30	0.64	(0.43–0.93)	0.69	(0.40–1.04)
Healthcare navigator during the diagnosis interval
No	96	1.00	(Reference)	1.00	(Reference)
Yes	119	0.96	(0.72–1.23)	0.93	(0.59–1.36)
Blocked to have doctor's appointment
No	127	1.00	(Reference)	1.00	(Reference)
Yes	74	0.80	(0.59–1.04)	0.88	(0.48–1.41)
Blocked to have appointment for biopsy taking
No	120	1.00	(Reference)	1.00	(Reference)
Yes	76	1.03	(0.75–1.37)	0.70	(0.36–1.14)
Blocked to have appointment for imaging
No	122	1.00	(Reference)	1.00	(Reference)
Yes	50	2.01	(1.41–2.70)	2.25	(1.40–3.34)
High cost of histology exam, transport to medical visits, and treatmentb	207	1.07	(1.01–1.14)	1.11	(0.95–1.26)
Appropriateness of doctors' attitudeb	207	1.13	(1.06–1.19)	1.07	(0.97–1.17)
Belief that the disease can be cured by a healer or maraboutb	207	1.28	(1.13–1.44)	1.43	(1.20–1.70)
Need additional support by selling property, borrowing and/or donationsb	207	1.14	(1.08–1.22)	0.91	(0.82–1.01)
Anomaly or symptoms are minor, hence no need for medical consultationb	207	1.03	(1.01–1.06)	0.81	(0.73–0.90)

CI: confidence interval; MAD, Moroccan dirham.

^aDiagnosis interval is defined as the time between the first consultation and diagnosis.

^bComponent scores created using principle component analysis with low scores indicating strongly agree and high scores strongly disagree.

Table 3 shows the factors associated with the time to initiate treatment following cancer diagnosis. The older age group had a treatment duration 2-fold longer than the younger age group, and patients who were currently married had a treatment initiation duration that was 21% longer. In contrast, patients with advanced stage cancer at diagnosis started treatment sooner than those with localised stage cancer, and those with a personal history of cancer started treatment sooner than those with no history. Patients in the highest household income category and those with easy access to the oncology centre (availability of public transportation or other means of transportation) also received treatment more rapidly than patients in other income and transportation categories. This study did not find any association between time to initiate treatment and gender, education level, health insurance coverage status, or place of stay during the ambulatory treatment. Again, the more a patient agreed with the statement “if an abnormality is not disturbing, there is no need to consult a doctor” or “the symptoms are benign and will clear”, the shorter was duration to start treatment (RR = 0.95; 95% CI = 0.91–1.00). The other financial, healthcare providers’ behaviours, and access to medical facilities factors were not associated with the duration of treatment interval.

Table 3.

Determinants of treatment interval^a (Marrakech, Casablanca, Rabat, and Fez centres data).

Characteristics	Patients assessed (n = 761)	Crude relative risk (95% CI)		Adjusted relative risk (95% CI)
Gender
Men	260	1.00	(Reference)	1.00	(Reference)
Women	501	0.77	(0.66–0.89)	1.02	(0.84–1.21)
Age group (yrs)
16–49	238	1.00	(Reference)	1.00	(Reference)
50–69	370	1.40	(1.16–1.62)	1.27	(1.06–1.52)
70+	151	2.13	(1.72–2.59)	1.98	(1.49–2.55)
Stage at diagnosis
I-II	251	1.00	(Reference)	1.00	(Reference)
III-IV	381	0.88	(0.75–1.03)	0.81	(0.67–0.96)
Unknown	130	1.30	(1.04–1.62)	1.27	(0.98–1.61)
Personal history of cancer
No	691	1.00	(Reference)	1.00	(Reference)
Yes	61	0.63	(0.46–0.83)	0.74	(0.53–0.99)
Education
No	468	1.00	(Reference)	1.00	(Reference)
Yes (primary, secondary, college)	266	0.85	(0.72–0.99)	0.88	(0.72–1.04)
Marital Status
Not currently married	219	1.00	(Reference)	1.00	(Reference)
Currently married	518	1.16	(0.98–1.36)	1.22	(0.98–1.47)
Total no. of children
0–1	183	1.00	(Reference)	1.00	(Reference)
2–4	300	1.10	(0.90–1.30)	0.93	(0.74–1.14)
≥ 5	233	1.71	(1.40–2.06)	1.07	(0.81–1.36)
Currently employed
Not	611	1.00	(Reference)	1.00	(Reference)
Yes	124	1.20	(0.97–1.46)	1.24	(1.10–1.39)
Personal history of cancer
No	691	1.00	(Reference)	1.00	(Reference)
Yes	61	0.63	(0.47–0.83)	0.74	(0.54–1.00)
Direct family history of cancer
No	660	1.00	(Reference)	1.00	(Reference)
Yes	88	1.10	(0.86–1.36)	1.04	(0.80–1.31)
Monthly household income (MAD)
None	188	1.00	(Reference)	1.00	(Reference)
<2000	311	1.16	(0.92–1.43)	0.95	(0.74–1.20)
2000–4000	169	0.78	(0.61–0.98)	0.81	(0.62–1.05)
> 4000	72	0.66	(0.48–0.87)	0.55	(0.39–0.76)
Public transportation close to home
No	252	1.00	(Reference)	1.00	(Reference)
Yes	499	1.09	(0.92–1.27)	0.81	(0.64–0.98)
Means to reach the oncology centre
Public transportation	559	1.00	(Reference)	1.00	(Reference)
Own means (car/motorbike/bicycle)	130	0.63	(0.47–0.83)	0.74	(0.54–1.00)
Relative or friends' car	52	0.51	(0.37–0.67)	0.48	(0.35–0.65)
Place of residence
Urban	437	1.00	(Reference)	1.00	(Reference)
Suburban	110	0.62	(0.49–0.76)	0.73	(0.56–0.92)
Rural	195	0.65	(0.54–0.77)	0.64	(0.49–0.82)
Place of stay during ambulatory treatment at the Cancer Centre
Own home in the same city as the Cancer Centre	244	1.00	(Reference)	1.00	(Reference)
Other place in the city of the Cancer Centre	173	0.62	(0.51–0.75)	0.84	(0.63–1.06)
Outside the city of the Cancer Centre	341	0.88	(0.74–1.04)	1.08	(0.86–1.37)
Type of health insurance
Basic health insurance (RAMED)	610	1.00	(Reference)	1.00	(Reference)
Civil servants and private insurance	114	0.95	(0.74–1.19)	0.78	(0.60–0.98)
None	27	1.41	(0.67–2.23)	0.84	(0.48–1.41)
Global cost of the disease (MAD)
≤5000	268	1.00	(Reference)	1.00	(Reference)
>5000	410	1.46	(1.22–1.73)	1.00	(0.76–1.25)
High cost of histology exam, transport to medical visits, and treatmentb	744	1.00	(0.96–1.04)	1.03	(0.99–1.08)
Appropriateness of doctors' attitudeb	744	1.00	(0.96–1.03)	0.97	(0.93–1.01)
Belief that the disease can be cured by a healer or maraboutb	744	1.00	(0.95–1.06)	1.03	(0.97–1.09)
Need additional support by selling property, borrowing and/or donationsb	744	0.95	(0.92–0.99)	0.96	(0.92–1.01)
Enough health centres and accommodations in city of the oncology centreb	744	1.09	(1.05–1.12)	1.04	(0.99–1.10)
Anomaly or symptoms are minor, hence no need for medical consultationb	744	0.92	(0.89–0.96)	0.95	(0.91–1.00)

CI: confidence interval; MAD, Moroccan dirham.

^aTreatment interval is defined as the time between diagnosis and treatment initiation.

^bComponent scores created using principle component analysis with low scores indicating strongly agree and high scores strongly disagree.

Discussion

The main determinants of the cancer patient pathway until treatment initiation were as follows: (1) access to diagnosis and medical facilities, thanks to public transportation or own transport means; (2) financial resources and its proxy (such as occupation, small families or having been to school); (3) a personal or familial history of cancer; (4) time availability (extensive family was a barrier for patient interval and having occupation for diagnosis and treatment intervals); (5) stage at diagnosis; (6) older age (as a facilitator until the diagnosis was made, than older age became a barrier to initiate treatment at the oncology centre); and (7) employee or private health insurance (with shorter patient and treatment intervals but longer diagnosis interval). Gender difference was observed only for the diagnosis interval, and the place of residence and place of stay during the ambulatory treatment did not have a clear impact on the intervals.

Findings from this study should be put within the context of Morocco. Morocco is a lower-middle-income country,¹² where poverty, illiteracy, poor health literacy, and poor cancer awareness are prevalent.¹³ The word ‘cancer’ is still taboo.¹⁴ As in many similar settings, the economic burden for cancer patients is immense, given the disproportions between treatment costs and household incomes; households are often pushed into poverty due to cancer diagnosis and cancer treatment.¹⁵ Patients must travel long distances to be diagnosed and to receive care, and transportation costs are often excessively expensive.^16,17 Another feature of Moroccan culture is its strong solidarity. It is mainly family-related, but help can come from the neighbourhood or even from private donors. The family management of healthcare is largely prevalent, reinforced by collective values that advocate family cohesion and solidarity.¹⁸ Since the first National Cancer Control Plan in 2010, access to cancer care of the population has improved. Implementation of the Medical Assistance Plan (RAMED or medical assistance scheme for those with no/limited incomes) provided free healthcare in the public health sector for deprived and vulnerable populations, although out-of-pocket expenses remain high when diagnosis procedures or treatments are not available at the oncology centre. In 2019, 45% of the population lived without any health insurance, out-of-pocket expenditure for health represented 47%; only 19% of the total population was covered by RAMED.^{19, 20, 21} In the present study, 81% of hospitalised patients were covered by RAMED, and 16% were covered by the employee or a private health insurance. This latter group represents wealthy and educated populations who promptly visit healthcare facilities following symptom recognition and who are often diagnosed and treated within the private sector.²² It should be noted that since the conduct of this study, Morocco has generalised the health insurance system (Assurance Maladie Obligatoire (AMO) or Compulsory Health Insurance). In July 2022, RAMED joined AMO. Morocco is making great efforts to implement Universal Health Coverage, with RAMED joining AMO being the first step.

Most of the population in Morocco use modern medicine and the proportion of individuals using traditional medicine is now low.²³ In the present study, 12% of patients declared having used traditional medicine before being admitted to the oncology centre. However, patients who disagreed with the statement that they “can be treated by marabout or traditional healers” were more likely to have a short diagnosis interval.

This study also found that having someone, very often a family member, to serve as a patient navigator to help coordinate and support the patient journey shortened (although not significantly) the diagnosis interval; this finding is consistent with previous studies.^{24, 25, 26}

The current study reported a significantly shorter treatment initiation among those with advanced stage cancer than among patients with a localised cancer. This has been described as the ‘sicker quicker’ effect; doctors tend to speed up the process to manage the more severe patients as quickly as possible. However, these patients generally have poorer outcomes as they are more likely to be diagnosed at a later stage.²⁷

Interestingly, the more patients disagreed with the statement that “the symptoms seem benign or not disturbing, there is no need to seek for a medical advice”, the longer the intervals for diagnosis, treatment, and overall. This counter-intuitive finding can be explained by the fact that the questionnaire was administered after hospitalisation at a time when cancer diagnosis had already been received. The patients understood, retrospectively, that they should have considered the situation seriously as the symptoms would not clear naturally and that they should have consulted a healthcare provider much faster. In a systematic review, it was reported that the presentation for medical advice occurs only when symptoms become incapacitating or impact normal activities; existence of a symptom alone cannot explain the patient interval.²⁸

Comparison of the treatment interval with international guidelines showed that Morocco was within the current Australian, Brazilian, Canadian, Colombian, UK, and US recommendations (≤28–31 days) but longer than the European Union recommendations (<15 working days [i.e., 21 days]). However, the primary care interval, the diagnosis interval, and the health service interval (sum of diagnosis and treatment intervals) were twice as long as the duration guidelines from Australia, the UK, and the World Health Organization (≤28–30 days) (Fig. 2).^{29, 30, 31, 32, 33, 34, 35, 36, 37, 38}

Fig. 2.

Comparison of intervals with international standards or timeframes (in days).

Delays in diagnosis and cancer treatment initiation are associated with a worse survival outcome and an increased mortality risk.^{39, 40, 41, 42, 43, 44} For instance, between 2004 and 2009 in the US, breast cancer treatment interval lengthened significantly and was associated with absolute increased risk of mortality ranging from 1.2% to 3.2% per week of delay.⁴¹ In addition to improving survival and quality of life, shortening diagnosis and treatment intervals is a matter of equity in giving the opportunity to all patients to be managed in the same effective way. It is also a measure of the quality of health services.⁴⁵ Indeed, in the US, time to diagnosis and to treatment of breast cancer has become the subject of accreditation promoted by quality-control agencies.⁴⁶ And, in the UK, the data on Cancer Waiting Times system are used to monitor performance against operational standards for cancer waiting times and aid service improvement.⁴⁷ Having to wait for a cancer diagnosis confirmation or to start cancer treatment is a stressful and worrisome experience for the patient and the family. However, the majority of countries, including high income countries, do not yet have defined standards for diagnosis and treatment intervals. Moreover, these targets are mainly focused on the waiting times to start treatment once the cancer is histologically confirmed, not from when the person first sees the general practitioner who suspects a cancer. Establishing target times for urgent suspected cancer referral or after an abnormal screening test, such as mammography, is essential. For instance, England has implemented the ‘Faster Diagnosis Standard’ with a target of no more than 28 days to confirm or refute a cancer diagnosis.

Indeed, while many countries present poor referral systems and inadequate patient-navigation systems, standard protocols for referrals should be developed between the primary care level and diagnosis and treatment facilities. Building local capacity and strengthening the referral system will help ensure timely and appropriate access to both local and tertiary care. Furthermore, patient navigation allows patients and carers to access available health services and reduce the number of health encounters and unnecessary steps. Navigators, such as community workers and health professionals, should be trained to assist patients in helping with appointment scheduling and coordination of care.⁴⁸ A prompt referral could be realised by scheduling rapid appointments for imaging, biopsy, endoscopy, colposcopy, biological exams, etc., grouped in one health facility and during a single visit but also by speeding up the waiting time for delivery of test results. Additionally, a multidisciplinary team approach (e.g., surgeons, medical oncologists, radiologists, nurses) should be in place to improve treatment plans and reduce duplication of care.⁴⁸

This study presents several limitations. The main limitation is the reliability of the date of symptom recognition from which several intervals were estimated. Best estimates for this date were made according to major personal or national events and in using information, if any, from the referral letter. Also, patients were interviewed just after the treatment initiation, minimising a potential recall bias. A second limitation is that some intervals were only based on data from patients in Casablanca, thus raising the question on the representativeness of the findings. Indeed, patients in Casablanca mainly came from the region and the southern part of Morocco. Finally, the study participants were mainly covered by the RAMED health insurance. Those covered by employee or private insurance were more likely to be treated in the private sector, where referral, biopsy, and treatment intervals are usually shorter.¹⁵ Therefore, the present study population is not representative of the Moroccan population in general but representative of the RAMED population who are treated in the public sector.

In conclusion, this study showed that in Morocco, the biopsy report interval and the treatment interval are within international standards; however, the diagnosis interval and the health services interval need to be improved. Current evidence suggests that interventions should firstly increase affordable and quality healthcare access and secondly, enhance cancer awareness within the population and the healthcare providers to productively speed up the referral once cancer is suspected; for instance, through implementation of Universal Health Coverage with a protection against catastrophic health expenditures.^21,49 It is recommended that health systems should define and regularly monitor interval targets through quantitative data in order to be more equitable, more effective, and ensure that cancer patients are treated within a defined timeframe.

Author Statements

Acknowledgements

The Authors are grateful to the patients and families who agreed to participate in the study. We are also grateful to the nurses and doctors who administered the questionnaire.

Ethics Approval

This study was approved by the IARC Ethics Committee and the Ethics Committee for Biomedical Research, Medical and Pharmacy School, Mohammed V – Souissi University, Rabat, Morocco.

Funding

Funding was received from the Lalla Salma Foundation, Morocco.

Competing Interests

The authors declare no competing interests.

Disclaimer

Where authors are identified as personnel of the International Agency for Research on Cancer/World Health Organization, the authors alone are responsible for the views expressed in this article, and they do not necessarily represent the decisions, policy, or views of the International Agency for Research on Cancer/World Health Organization.

Appendix A. Supplementary Data

The following are the Supplementary data to this article: