On fevers in primary care- and the lessons we learn for health systems strengthening.
TS : One of the commonest problems that we see in primary care and yet one of the most difficult is fevers. What is primary level care worth if we cannot manage a simple fever. But fevers are seldom simple. Nor can we send every person presenting with fever to tertiary care because it is so difficult to access, but also because fevers are so often self-limiting. When policy talks of shifting from selective to comprehensive primary care, it is largely referring to a few non-communicable diseases (NCDs). But the earlier package of selective primary care was restricted to less than one fifth of all infections. That package never did address the majority of fevers, and even now the transition to comprehensive primary health care also underplays this. So what’s your take on that? How do we approach this?
YJ: As you rightly said, fever is the commonest symptom with which people present to a healthcare provider in this country, and this may be true across levels, from ASHA workers to the tertiary hospital. Fever, as we know, is a symptom that is most commonly associated with infections, which could be bacterial or viral or caused by other micro-organisms. Other causes of fever are a very small fraction, like the occasional instance when a cancer or autoimmune disease presents with fever. But largely, if a patient presents with fever, it is an infection, a communicable disease.
Most infections are self-limiting, but a significant proportion, develop complications and result in death. Overall, fever-related deaths remain high, and this continues to challenge all of us in public health.
Fevers can present with a localising symptom which indicates the organ that is infected, but just as often it can present as an undifferentiated fever without any localisation to any organ. A fever along with cough with sputum would indicate a respiratory disease like pneumonia. If accompanied by diarrhoea it is gastrointestinal, if with burning urination, then it is a urinary tract infection, and if a person is having an ear pain, it is probably an ear infection and so on.
The greater problem is with undifferentiated fever, typically one that has lasted more than two or three days, but less than two weeks at the time of presentation, and without any localising symptoms. Such a fever can be benign to begin with, like a throat infection, but it could later turn serious. Also, such fevers could be undifferentiated initially and later develop a localising symptom. Undifferentiated fevers often occur as a cluster of symptoms, which we call a syndrome: like fever with a rash, with enlarged lymph nodes or with jaundice. For each such syndrome one has to consider 5–10 causes. In a primary care setting, the challenge is to find the simplest and most cost-effective and reliable way to offer stabilising care prior to referral for diagnosis and treatment.
TS: So, what are the usual causes of fever across the country? Approximately what proportion of deaths would it be causing?
YJ: Earlier it used to be malaria. Now typhoid fever, also known as enteric fever, is one of the largest unsolved problems of febrile illness in the country. The other common fevers include dengue, chikungunya and zika which are caused by mosquito borne viruses. Within the last decade scrub typhus which was earlier endemic in a few pockets has become very widely prevalent. More rare and more localised are viruses like Nipah, Chandipura, and West Nile, Kyasanur Forest Disease(KFD). And let us not forget about leptospirosis, which is a very poor person’s illness, and an illness that is associated with flooding in urban and rural areas as another common cause.
The other common infection presenting as fever is the flu, that is an influenza virus. The patient may have some respiratory symptoms which may or may not worsen. Covid-19 too is around, and we still get sporadic cases.
Overall, te chances of any of these illnesses becoming serious enough to kill would be less than 1% in almost each of those illnesses, except maybe malaria, typhoid and leptospirosis, where mortality is higher.
TS: In practice in our outpatient departments and general practice, we often put down a short duration fever as viral fever, and we investigate it only if it persists beyond four or five days. Now what is this viral fever, and does such a practice make sense?
YJ: This rule of thumb may not work in every situation. I think it’s very contextual. Geography, seasonality, and local, recent epidemiology of infections must guide the decision. Infections like falciparum malaria, scrub, Chandipura, or Japanese encephalitis (JE) can become serious and cause unconsciousness and death in a short period. So a four-day window period does not always work, unless you have some respiratory symptoms. Respiratory viruses like rhinovirus, respiratory syncytial virus, or influenza occur in certain months, and they are usually minor and self-limiting except in the very young or very old.
But with most other diseases it is important to identify those who are likely to turn sick, and begin investigations even earlier. This is done by identifying some red-flag signs, or because we suspect something, based on the current prevailing infections in that locality, and not just wait and watch for the first four days to pass.
TS: Why is it important to diagnose the causative agent? Most GPs would probably treat empirically―with paracetamol and an antibiotic―and this has become the public expectation also. Waiting for diagnostic reports to come in is not the preferred practice.
YJ: One reason for insisting on diagnosis of the cause in any acute undifferentiated fever is to pick up those illnesses, such as scrub, typhoid etc., that can turn serious.
Another is to resist the knee jerk tendency for every practising health care provider to prescribe an antibiotic almost as an antipyretic to bring the fever down, assuming that it will do no harm, but it may do good. This misuse of antibiotics has led to the burgeoning problem of antimicrobial resistance. At the moment the resistance situation is completely unmanaged, but there is callousness about it. This is a national shame and a global problem. And this is the second equally important reason to diagnose the cause of infection before we prescribe anti-microbials. Unfortunately, the availability of necessary diagnostics is far from adequate. We have invested in diagnostics for malaria and now for dengue, but not for many others like scrub typhus which is now one of the commonest causes of fever-related complications and deaths in the country, across regions.
TS: But what are the protocols in place for fevers of different duration? And the diagnostics? Are the guidelines in place? Do providers have the necessary clarity?
YJ: There is a protocol [1] [2] for what should be asked for even on the first day of fever. And whenever they examine the person, they should look for red flag symptoms like altered consciousness, vomiting, seizures, or low oxygen levels, and if any of those symptoms are present, they’re supposed to refer the patient to a higher level facility. They are also required to enter the data of such patients into a web-portal―the Integrated Health Information Portal, (IHIP)― which is the main tool of disease surveillance. This IHIP portal[3] calls for daily reporting. The frontline worker’s diagnosis is called a suspected case, and if it reaches the doctor it is a probable/presumptive diagnosis. The laboratory provides a confirmed diagnosis.
In addition, in many places primary centres have double diagnostic kits, which test for both dengue and for malaria together, and the recommendation in all malarial areas is to do the double test as soon as the patient presents with fever. No anti-malarial is to be prescribed unless there is a positive test, and no anti-microbial either. This is repeated on the fourth day if the fever persists. And then to refer to a physician at a higher-level institution to diagnose other illnesses like enteric fever. The system does not recommend that the ASHA administers antibiotics for enteric fever or suspected enteric fever.
And so, if you were asking what one should be doing as a public health measure, my emphasis is to invest in getting diagnostic facilities accessible to each health provider. There is now a prescribed list of diagnostics for every level of the public primary healthcare system[4]. But this needs to be enhanced and ensured.
This may still miss out on other care providers such as informal practitioners, faith healers, AYUSH physicians and even self-treatment. A significant number of acute fevers stay off the radar since these health providers don’t have access to free diagnostics.
TS: What diagnostic kits are you suggesting? Other than dengue and malaria.
YJ: Definitely for scrub-typhus, where rapid diagnostic kits are available. Then a greater access to blood culture. This may require transportation of the blood sample to a hub. The Widal test for typhoid is inadequate and, in fact, should not be used at all.
TS: I see that you have not mentioned blood counts, or ESR or platelets, at least for dengue.
YJ : These are not recommended for investigation in a routine fever. They could be part of the work-up if an in- patient. workup in a patient with persistent fever over three weeks, or if there is some specific indication for the same. Platelet counts in dengue needs to be done only if there is bleeding. Platelet counts can be low in patients with most of the other causes of fever we talked about, scrub, leptospirosis, Zika etc. So, it does not help in diagnosing or even following up people with dengue. What is required is to monitor the fever pattern, and pulse rate, and whether the patient is hydrated.
TS: Coming to the specific issue of dengue, which is arguably the most common cause. Why have we not developed herd immunity? Doesn’t an attack of dengue provide life long immunity? We see this in chikungunya – the outbreak ceases as herd immunity rises. It returns after 15 or 20 years, surges and then subsides.
YJ: Dengue has four serotypes. When first infected with one strain, the proportion who develop a serious illness is very low, less than 1%. But when the patient gets subsequently infected with a second strain, say a couple of seasons later, they are likely to have a worse disease. The occurrence of a previous infection increases the chance of developing dengue shock syndrome or a dengue haemorrhagic fever in the second episode. And since each outbreak is with a different strain, the proportion of complications that happen with dengue will increase with time. And while herd immunity prevents an epidemic spread, it will still remain endemic. Dengue is probably the new ‘malaria’ for us in the country, with every year leading to successively increasing numbers of people with infection and with death. It began as an urban illness, but has now become more and more common in rural areas. And the longevity of the Aedes mosquito that transits it has increased because of higher summer temperatures. So, dengue is happening in more months of the year than before. Water scarcity adds to this problem, as there is greater storage of water now in both urban and rural areas and the mosquito proliferates in stored water. So, we have a situation now where the mosquito burden is increasing, and the number of infections due to dengue as well as complications of dengue, and dengue-related deaths, have increased phenomenally in the last five years, across both rural and urban areas.
TS: And, what is the state of the public health response to this growing problem?
YJ: We have a good dengue test which is now available in most primary health centres. And these have been supplied to frontline workers like ASHAs also in certain parts of the country. The test is reasonably sensitive, even in the first three days of infection. The challenge is that even after diagnosis and first contact care, we need to subsequently monitor these patients. Admission is not required. They can be monitored at home also.
Because of prevailing misinformation, the diagnosis causes fear, and doctors prescribe―and patients submit themselves to―useless blood tests and irrational therapies. In preventing breeding of Aedes much has been done, but much more requires to be done, and consistently, with much greater community participation. That would prevent not only dengue but also other mosquito-borne disease like chikungunya.
TS: And what is the situation regarding Chikungunya?
YJ: Chikungunya too has increased remarkably. Initially a South Indian infection, it is now prevalent across India. Year 2024 was bad for this disease, but 2025 was relatively quiet. Since the disease follows a biennial pattern, it may be bad this year. One specific feature of Chikungunya is that it does not kill as frequently as dengue does, but it results in much disability. A high proportion of adults, above 20 percent, develop disabling joint pains, which may last even a year or two. And it does not give life-long immunity. It can recur. There is no rapid diagnostic kit for Chikungunya and diagnosis is clinical. Only a few cases are confirmed by serological testing.
TS: Now, where are we with malaria? Are we on course to eliminate the disease? And do we need such high levels of annual blood examination, given such low incidence?
YJ: In the last seven years, there has been a successful reduction in the incidence of malaria. This is largely due to reduction of falciparum rates in the high burden districts of the main high-endemic states Chhattisgarh, Odisha, Jharkhand, Madhya Pradesh, and the North-East. We did some things well, such as giving rapid diagnostic kits and artesunate combination therapies to our ASHAs, and effective distribution of long-lasting insecticidal nets (LLINs). But then we went into over-reach and decided we will eliminate it by 2030- which meant that by 2027, we would have zero indigenous cases. Unfortunately, malaria has made a comeback in the last two years, with a fairly dramatic increase in the number of cases in 2024 and 2025 in Odisha. There have been sporadic increases in other states of the country, but the numbers are still much lower than 2016.
It is the usual lowering of the guard that happens when elimination targets are approached. The LLINs were last issued in 2019, and given that they have a life-span of four years, they should have been re-issued in last two years. They have not been. There are interruptions in supply of artesunate and Rapid Diagnostic Kits (RDKs) too. So even falciparum may re-emerge.
At the same time, we note that vivax malaria did not come down significantly. In some cities like Mumbai and Delhi it has increased. Vivax is a challenge because cases are often asymptomatic and these patients act as reservoirs of infection. Community-level screening is difficult to organise, but must be done where vivax is being reported. To prevent the carrier state from developing further, those infected with vivax will need to be given primaquine for 14 days, and this has been a challenge to achieve. And finally, the vector Anopheles stephensi is now much more prevalent. So not only are we worried about missing the 2030 target, we need to worry about a possible resurgence unless we address these issues.
TS: Okay. That’s comprehensive on malaria. Can I draw your attention to typhoid?
YJ: I think we are in deep trouble with typhoid. Typhoid kills, and typhoid is seeing a rapid emergence of anti-microbial resistance, increasing its propensity to be fatal.
The big problem with typhoid is that both the main diagnostics available―Widal test and Typhi dot[1],[2], ―have low diagnostic accuracy, especially in the first four days. Blood cultures are still the standard, but that is not easy to access. Widal would be more reliable if there were two samples drawn one week apart. But that is seldom done. I think the priority should be to develop a rapid diagnostic kit and make it widely available. But I don’t know of anything that is even on the horizon, that will replace current diagnostics.
Poor reliability of diagnostics leads to presumptive prescription of antibiotics, and the misuse of antibiotics in such settings has led to typhoid becoming resistant to most antibiotics (the quinolones like ciprofloxacin, co-trimoxazole, amoxicillin, chloramphenicol etc.) Now third generation cephalosporins are the mainstay, and resistance is developing even towards them. So drugs like carbapenems are coming into use, which is not a good solution at the public health level, being an expensive and a more difficult drug to administer.
So with regard to typhoid, we are really in a big soup. And you know, typhoid is one illness that binds the entire country. I don’t know one state in the country where typhoid is not common.
TS: When guiding students doing their internships, I was always amazed to find that the disease surveillance portal reports high incidences of typhoid, but no public health action follows. This is true for all fevers, but most problematic in the context of typhoid- a food and waterborne disease that admits of multiple modes of preventive public health action. This is in contrast to high income countries where a single case of typhoid sets off a public health inquiry and necessary action.
YJ: I am not sure, butwould speculate that since we do not have a reliable diagnostic test, we do not take the reports seriously. And the lack of knowledge on what requires to be done. When you are clueless, you tend to dismiss it. And as a public health practitioner. I’m embarrassed, even ashamed, at our neglect of one of the commonest illnesses that we encounter. If you go to the medicine and paediatric wards in most cities, a significant proportion of beds will be of patients with longer duration of fevers, and many of these would be drug resistant typhoid. I think this is blindness in plain sight.
TS: On scrub typhus, why has it suddenly become so widespread, and what are the public health measures against it?
YJ: Scrub typhus has been around since the ’60s, but it is only in the last two decades that it has become so widespread in most East and South East Asian countries. It is now one of the commonest causes of fever related complications- especially the acute encephalitic syndrome (AES). Till about 1995, Japanese encephalitis was the most common cause of AES. Now it is scrub, even in Gorakhpur, where traditionally we heard that epidemics of JE were regularly seen.
Its called a typhus since it is often associated with a rash. In about 60% of the affected persons, there is a black necrotic scar, known as eschar, which looks like a lit cigarette butt has touched your skin. That is where the mite bit. The standard blood test is known as the Weil-Felix test. There are now some rapid diagnostic tests available, but not in the public system. Diagnosis by ELISA and PCR test are most reliable, but except for medical colleges, this test is seldom available. This fever can present only as complications reflecting the organs affected-, pneumonia, jaundice, renal failure, unconsciousness. And duration is variable from 3 or 4 days to over 30 days.
As of now treatment is effective if diagnosed early,, and the antibiotics that act against it best are doxycycline and then azithromycin. Both, common first line antibiotics, and there is as yet no evidence of resistance to these.
There is limited understanding of the public health aspect. It is usually acquired from exposure to vegetation, from walking on grass, on the edge of fields or water bodies, in forests and so on Hence the name scrub.-. Mites inhabit these locations. There is a need to integrate this into the vector-borne disease control programme.
TS: Part of the problem with integration, I think, is that so little is know about the vector dynamics. The mite has four life stages―egg, larval, pupa and adult―and of these it is only the larval stage, also known as chiggers that is infective, since this is the only time the larva feeds on vertebrate tissue fluid. The infective micro-organism is a bacterium of the Rickettsiaceae family- called “Orientia tsutsugamushi”. Mites get it from infected rodents and other small animals, where there is an established cycle with no obvious disease. It only incidentally passed on to humans in whom it causes the disease. This is more likely to happen where vegetation that harbours rodents is plentiful and there is human movement in this area. Mites transmit the disease to vertebrates including humans only in the larval stage. But mites transmit the bacteria to its own next generation in its reproductive cycle. It’s called transovarian transmission. But why did it suddenly break out? One common attribution is to climate change and loss of bio-diversity[1]. Preventive measures are also almost non-existent. The best bet as of now is personal prophylaxis as practiced for all other vector-borne disease, where those likely to be exposed need to wear clothes that cover the body, use insect-repellent creams, wash-up after a visit to the farm or forest and such. And there is no work on a vaccine.
YJ: I would really argue strongly for including scrub typhus in the vector-borne disease control programme.
TS: We have talked of a few vector-borne diseases that present as fever, but can turn fatal. But there are others which are highly localised: notably Nipah, Chandipura, KFD, West Nile and Zika virus. Two other localised disease outbreaks that are rare but deadly are amoebic encephalitis and meningococcal meningitis
YJ: Well, these are rarer, and localised, but taken together they are a significant threat, and every one of them can potentially break their current boundaries. They share common determinants like being mosquito borne, or being related to increased warmth of pond waters (for amoebic encephalitis) and floodings. They all require good laboratory support to diagnose and good tertiary care linkages to treat. If we do better with the more common communicable diseases like typhoid and cholera, we will be better placed to deal with these too. The reason why Kerala reports these new infections like Nipah and amoebic encephalitis much more is precisely because they have good primary care and competent public tertiary care linked to better laboratory support. Chandipura was diagnosed in Gujarat with laboratory diagnostics done at NIV Pune. The rapid increase in such illnesses is a challenge, but the good news is that 150 virus research and diagnostic laboratories have been mandated across the country. These have taken root in the southern states, but most northern states are still dependent on ICMR and NIV teams.
TS: Another major problem is leptospirosis.
YJ: Yes, Leptospirosis, once characterised by sporadic, seasonal localised outbreaks is now endemic, yearlong and widespread. Much of this is associated with flooding, especially when such flooding causes sewage to come out into the open air. Till some years ago, Leptospirosis was still a rural illness, seen by the poorest farmers where, when they were sitting, you know, drenched in rice fields, paddy fields. But now it has become an urban phenomenon. It has been reported from urban Kerala, from Tamil Nadu, from Andhra Pradesh and from Mumbai, most often associated with floodings, where sewage, rodents, and human bodies come into close contact[1]. In these areas, anyone standing in flooded water for over an hour runs the risk of getting something like leptospirosis. These diseases used to be thought of as exotic, but we are probably diagnosing them a little better now and as part of the routine. What all these challenges demonstrate is the need for a coordinated public health response, rather than merely reactionary to a specific outbreak.
TS: What about new technologies, vaccines and diagnostics. In the last conversation we talked about HPV vaccines that prevent cervical cancer, but what about vaccines for these common and widespread scourges like dengue and typhoid?
YJ: A dengue vaccine is available. And there is a strong case for the typhoid vaccine, because it’s an illness that has a poor diagnostic test and because the treatment has become difficult due to anti-microbial resistance. The good news is that the revised new typhoid vaccine is over 85 to 90 percent effective, and its effectiveness lasts about five years. I would strongly recommend the national programme to take it on. Bangladesh has already deployed a nation-wide typhoid vaccination programme with positive outcomes[1]. Similarly, the dengue national programme should also include vaccination. There is a two-dose Japanese vaccine, the Kadenga vaccine, that is available, and there is an indigenous multi-valent single dose vaccine under development. Brazil and other Caribbean countries have gone in for a two-dose dengue vaccine.
TS: Another big question: What is the progress with disease surveillance, and how is the IHIP portal, on which this is based, faring?
YJ: Maybe we can explore this in another conversation. It has come a long way, but I think it still falls short in many critical areas. There are technical issues related to digitisation and portal functionality, there are data related issues, related to the considerable amounts of data that are to be entered, and there are issues related to use of the information. I think the IHIP functions reasonably where public health systems are functional.
TS: Finally―and we cannot close without this―where are we with the surveillance and response to Influenza-like illnesses, and to Covid-19? To what extent are these persistent problems? After 2012, we did establish a fairly robust system for surveillance of influenza-like illness (ILI) and Severe Acute Respiratory Illness (SARI). We sidelined it during the covid-19 pandemic, where Covid-19surveillance arose as a parallel system. But hopefully it is back and integrated with Covid-19 surveillance. And what about the introduction of anti-flu vaccination for the elderly?
YJ: I am not sure how the flu-surveillance system is now. Unlike earlier, there are very few reviews and monitoring mission reports. I can say that there is a need to increase our diagnostic abilities at the secondary care level of the district hospitals and higher. We do not have flu alerts or advisories on the latest infection patterns and threats. So clearly there is some distance to go.
As for anti-flu vaccines, we may not be able to afford the vaccine for everyone, like many developed countries which offer it to all the entire age groups. But we could offer it to those with chronic morbidities, as well as the elderly above 60 years of age.
YJ and TS: We have explored many themes. Let us together try to draw out the main lessons from this conversation – the five take-home messages so to speak.
Our first message: In this policy thrust of moving from selective to comprehensive primary health care, the emphasis has been on the inclusion of non-communicable diseases. While that remains important, this discussion on fevers highlights that primary care for communicable diseases is still very incomplete. We also need to disaggregate mortality data for fever-related deaths, to assess the continuing burden of infectious diseases.
Our second message would be the need to ensure that primary care providers are able to determine the cause of fever and provide cause-specific treatment in most of the fevers that present to them, and do so as soon as they present. This will save lives, and prevent the rise of anti-microbial resistance. This, in turn, requires much better two-way linkages with diagnostic laboratory support and better availability of point-of-care diagnostics. We also need to provide updated protocols on fever management to our frontline workers and train them to be able to use these effectively.
Our third message would be to leverage the disease surveillance/IHIP system with good laboratory support, possibly through regional antimicrobial sensitivity data, to generate periodic district-specific reports on the pattern of fevers by cause, and in the form of district level advisories to healthcare providers at all levels, both public and private. This would improve the management of fevers.
Our fourth message: Much better public health preventive action in the form of better management of water and sanitation, vector-control, and zoonosis, which must extend to these neglected causes of fevers. Even for the national disease control programmes like tuberculosis and malaria, public health action is weak, but in situations like typhoid and scrub typhus it is almost a non-starter.
And finally, we draw attention to the need for technology development for the common fevers. We need better point-of-care diagnostics, better therapeutics and better vaccines all of which are under-invested in. Since these diseases, though very widespread, remain diseases of the poor, relevant technology development has been neglected.
YJ: I would just add to that one immediate area for action is the need to decentralise, and democratise access to quality diagnostic laboratories. In the trajectory of development of decentralised health systems, we tend to focus on drugs, then human resources with clinical skills in diagnosis, then better outreach with community health workers, mid-level workers, doctors etc. Diagnostics comes in last, and we have been held back both by the delay in development and deployment of diagnostic technology.
We may have made some progress in prevention, and with decentralisation and community participation its effectiveness can increase further. But we have been held back by insufficient technology, particularly diagnostic technology. We have not democratised this enough. And we have also been slow in vaccine development and deployment that is responsive to our needs, as we noted with respect to the typhoid vaccine. But the larger point is that we are associating universal healthcare only with non-communicable diseases. This conversation brings back the extensive and emerging role of communicable diseases.
(We acknowledge the inputs from Dr Vasundhara Rangaswamy. And the assistance of Dr Arun Krishna in recording and transcription and Dr Gayatri Sabherwal in editing)
Note: This is the 32nd conversation in the series. Readers can enter into the conversation by providing their feedback at the end of this article on the website where it is posted, or on any of the social media platforms where it is circulated.
To access the earlier conversations and other curated information on health policy and health systems strengthening please visit the website : https://rthresources.in/ or https://rthresources.in/conversations-on-health-policy/
