GERIATRIC Medicine THESIS TOPICS FOR MD/DNB

A geriatrics protocol is judged on internal consistency: the research question, objectives, methodology and statistical plan must all describe the same study. The primary objective should be a single measurable endpoint — one prevalence, one association, one comparison between two defined groups — with everything else demoted to secondary objectives.

Name every instrument, and understand that the choice is the study. Almost nothing on this list is measured with a machine; it is measured with a scale, and different scales measuring the same construct do not identify the same people. Frailty assessed by a physical phenotype requiring grip strength and gait speed, by a clinician-rated judgement scale, and by a five-item questionnaire will produce materially different prevalence figures from one population, and all three are legitimate. So state which instrument, which version, how many items, the scoring rule, and the cut-off defining the abnormal category, with the reference from which the cut-off is taken. Do this for every construct in the study: cognition, functional dependence, instrumental activities, nutrition, depression, sleep quality, loneliness, social support, quality of life, caregiver burden, fear of falling and pain. A protocol saying "frailty was assessed" or "quality of life was measured" has not described a method, and it is the commonest reason a geriatrics protocol comes back from review.

Deal with translation and permission before submission, not after. Two practical points that catch residents late. First, an instrument administered in Marathi, Hindi or any language other than the one it was written in must be a properly validated translation, cited as such — not a version produced by the resident in the clinic the week before. Where no validated translation exists, the protocol must describe a forward-and-back translation with expert review and pre-testing, and that becomes a component of the study rather than a footnote. Second, several of the instruments in common geriatric use are not free to use: some cognitive screens, several quality-of-life measures, the standard caregiver burden inventory and some nutritional tools require permission, registration or a licence. Confirm the terms and state in the protocol that permission has been obtained, because a thesis is one thing and a journal submission asking for the permission letter is another.

Address the education and literacy problem directly. Cognitive screening instruments were developed in literate populations and their scores depend heavily on schooling. Applying a standard cut-off to an older Indian population with limited formal education manufactures cognitive impairment in large numbers of people who have none. State the education-adjusted cut-offs you are using, or use an instrument specifically adapted for low-literacy settings, and describe how illiterate participants are handled for items requiring reading or writing. The topic examining educational status against cognitive function is partly measuring this artefact, and its discussion section must say so rather than reporting the association as a pure biological finding.

Define the setting honestly. "Community-dwelling" appears in several titles here and it means older people sampled where they live, not older people who happened to attend a hospital outpatient department. Clinic attenders are a selected group: sicker, more mobile than the housebound, and dependent on having someone to bring them. If your sampling frame is the outpatient register, say so and title the study accordingly. If a topic compares community-dwelling with institutionalised participants, describe both sampling frames separately and describe the institutions, since Indian old age homes range from charitable shelters to paid retirement housing and the residents are not comparable across those categories.

State the age threshold and how age was ascertained. Indian policy and most Indian research define older adults from sixty years, while much of the international literature uses sixty-five; declare which you are using and cite it, because it determines whether your prevalence figures can be compared with anything. Say also how age was established, since many older Indians have no birth record and reported ages cluster heavily on round numbers.

Say who the informant is. Several constructs here can be reported by the participant or by a family member, and the two disagree systematically — caregivers tend to report more dependence than the older person does. Fix the rule: self-report throughout, proxy report where cognition is impaired below a stated threshold, or both with the difference examined. For the medication topics, state whether counts come from prescriptions, from the pill bag brought to clinic, or from recall, and whether over-the-counter drugs, supplements and traditional preparations are included, since their exclusion substantially undercounts medication burden in Indian practice.

Name the criteria sets. Potentially inappropriate medication use requires an explicit list with its edition, and the widely used international lists include drugs unavailable in India while omitting drugs commonly prescribed here — say which list you apply and acknowledge the mismatch. Sarcopenia requires a named consensus definition with muscle mass, strength and performance components; use the Asian criteria with their cut-offs rather than European ones, and state the method for muscle mass, the dynamometer used, the position and number of grip attempts, and the gait speed course length. Delirium requires a named validated instrument, and a deeply sedated or unarousable patient must be recorded as unassessable rather than delirium-free.

A geriatrics synopsis is the condensed document of two to four pages — title, introduction, aim and objectives, brief methodology, sample size and references — submitted for registration of the dissertation topic. The geriatrics protocol is the expanded version of twelve to twenty pages carrying the full review of literature, detailed methodology including every instrument with its version, translation, scoring and cut-off, the sampling frame, statistical plan, study timeline and annexures.

The annexures matter more in this speciality than in most, because the instruments are the methods. Include the full text of each scale in the language of administration, or where licensing prevents reproduction, cite the source and state that permission has been obtained. Include the information sheet and consent form in the local language, written to be read aloud, since a substantial proportion of participants will not be able to read them. Include a separate consent form for caregivers in the caregiver burden topics, because the caregiver is a research participant in their own right and needs their own consent, not the patient's. And include a short written pathway describing what happens when a participant screens positive on the depression, cognition or nutrition instruments.

In practice the synopsis is extracted from the protocol rather than written separately, which is faster and produces a more coherent document. Check your university's prescribed proforma before submission, since rejections on formatting grounds are common and entirely avoidable.

Match the formula to the design. Prevalence topics — frailty, cognitive impairment, falls, malnutrition, sarcopenia, incontinence, depression, delirium — use a single-proportion calculation with the expected prevalence taken from a cited Indian study rather than a Western one, since the figures differ substantially. Comparative topics between two defined groups need a two-proportion or two-mean calculation with both expected values referenced. Correlation topics such as grip strength against functional mobility, or dependence severity against caregiver burden, use the expected correlation coefficient. Where the study is a community survey using cluster or multistage sampling rather than a simple random sample, apply a design effect and say so — omitting it understates the required sample considerably.

Frame everything as association, not effect, and say why. This deserves its own paragraph because it runs through the entire list. Frailty and depression, sarcopenia and dependence, loneliness and depressive symptoms, chronic pain and quality of life, polypharmacy and falls: every one of these pairs is plausibly bidirectional, and a study measuring both at a single visit cannot establish which came first. Polypharmacy is the clearest illustration — drugs associated with falls are frequently prescribed for the very conditions that cause falls, so an observed association partly reflects the indication rather than the medication. Write the objective as association, avoid causal verbs in the results, and state the direction problem in the discussion plan rather than leaving it to the examiner to raise.

Falls are recalled, and recall decays. A cross-sectional study asking about falls in the preceding twelve months will undercount, because minor falls without injury are forgotten within months and recall is poorer in exactly the cognitively impaired participants most likely to have fallen. State the recall period, state who was asked, and report the under-ascertainment as a limitation with its likely direction. The same caution applies to self-reported medication use and to duration of symptoms.

Name the tests. Proportions use the chi-squared test with Fisher's exact test for sparse cells. Continuous variables use the independent t-test or Mann-Whitney U test with a stated normality test; scale scores are frequently skewed and bounded, so plan for medians. Ordinal outcomes — frailty categories, dependence grades, clinical stages, burden bands — are analysed with rank-based tests and correlated using Spearman, not converted into means. Comparison of a scale score across three or more groups uses one-way analysis of variance or the Kruskal-Wallis test with a stated post-hoc correction.

Two disciplines worth imposing on yourself. First, these topics invite multivariable logistic regression, and the events-per-variable rule applies: roughly ten outcome events per predictor entered. A study of one hundred and twenty participants with thirty frail cases cannot support a model with twelve predictors, however tempting the table. Decide the predictor list in the protocol and justify its length. Second, when several instruments are administered to the same participants, the temptation is to cross-tabulate everything and report whatever reaches significance. Declare the primary comparison in advance, treat the rest as exploratory and label them so, and state a correction if you intend many comparisons.

Where you administer the instruments yourself, report reliability: have a second assessor independently rate a subset, at least ten per cent, and give the kappa for categorical classifications and the intraclass correlation coefficient for scale scores.

Start with where your older patients actually are. A dedicated geriatric clinic running two or three days a week with thirty to fifty attenders a session will support any of the outpatient prevalence topics comfortably. Where no separate clinic exists, the older population is spread across general medicine outpatients and wards, and you will need a defined sampling frame rather than convenience recruitment — decide that before registration, not during collection.

Then count the time each assessment takes. This is the constraint that catches residents in this speciality, because the instruments are administered by you, one participant at a time. A comprehensive assessment covering frailty, cognition, function, nutrition and mood takes forty-five minutes to an hour with an older person who is hard of hearing and accompanied by a family member with questions. Three hundred participants at that rate is three hundred hours of clinic time. Either narrow the instrument set to what the primary objective actually needs, or reduce the sample to something honest.

Finally, check three specific things before committing: whether a hand dynamometer is available and calibrated, since several frailty and sarcopenia topics are impossible without one; whether body composition measurement is accessible at all, since sarcopenia diagnosis strictly requires a muscle mass estimate; and whether the instruments you intend to use are free to administer or need licensing. Any of the three can end a topic after registration.

Cross-sectional observational and analytical designs dominate: prevalence of a geriatric syndrome in a defined population, association between two assessed constructs, comparison between two defined groups such as frail and non-frail or hospitalised and ambulatory, and profile studies describing a clinic population. Instrument-comparison studies, in which two screening tools are applied to the same participants and their agreement or relative yield reported, are well established and often produce a cleaner result than another prevalence estimate.

Knowledge and attitude surveys of older adults or caregivers are accepted and require a pre-tested or validated questionnaire. Caregiver studies, in which the caregiver rather than the patient is the participant, are equally acceptable and increasingly popular. Longitudinal designs following participants for incident falls, functional decline or mortality are the scientifically stronger option but rarely fit a three-year programme with an eighteen-month collection window; interventional designs are largely outside postgraduate scope.

Bring three to five shortlisted titles rather than one, since guides frequently rule out a topic on grounds you could not have known — a departmental study already running, an overlapping subject held by a senior resident, a community outreach programme ending, an instrument the department has decided to standardise on.

Settle five things in that meeting: the sampling frame and whether the study is genuinely community-based or clinic-based, which instruments the department uses and whether validated local-language versions exist, how long one full assessment will take and therefore what sample is realistic, who will administer the instruments and whether a second assessor is available for reliability, and which journal the eventual paper is aimed at. Where the topic needs another department — psychiatry for the cognitive and depression work, physiotherapy for mobility measures, radiology for bone densitometry, community medicine for field access — secure that cooperation in writing rather than assuming goodwill will hold.

Rarely, and it is worth being clear about why, because residents attracted by the speed of a record-based study often discover the problem too late. The central variables in this list — frailty status, cognitive score, functional dependence, nutritional risk, mood, quality of life, caregiver burden — are produced by administering an instrument to a person. They are not in the case sheet. No amount of chart review recovers a frailty score that was never measured, and reconstructing one from recorded clinical details is not the same instrument and cannot be reported as though it were.

A small number of topics here are genuinely record-friendly: the comorbidity profile of older admissions, prevalence and pattern of electrolyte abnormalities, anaemia among hospitalised older adults, and potentially inappropriate medication use where the prescription record is complete. For those, state the period searched, how cases were identified, and the completeness of retrieval, and note two specific cautions — that discharge diagnoses systematically under-record geriatric syndromes such as delirium, incontinence and falls, so a record-based prevalence is a floor rather than an estimate; and that medication lists in case sheets usually omit over-the-counter, supplement and traditional preparations entirely.

Everything else on this list needs prospective administration. Plan for it in the timeline rather than hoping the archive will do the work.

The synopsis is the condensed two to four page document submitted for topic registration. The protocol is the full document of twelve to twenty pages containing the detailed review of literature, methodology with every instrument named by version, translation, scoring and cut-off, the sampling frame, statistical plan, timeline and annexures including the scales themselves in the language of administration. The synopsis is normally extracted from the completed protocol.

Full institutional ethics committee approval before data collection. Five points need explicit attention in this speciality, and committees reviewing geriatric protocols look for them.

Capacity and consent. Older adults are not a vulnerable group by virtue of age, and the protocol should not treat them as though they were — a competent eighty-year-old consents for herself. But the dementia, delirium and severe cognitive impairment topics will enrol participants who cannot. State how capacity is assessed, who may consent on their behalf under the term your committee uses, and that the participant's own assent is sought and their refusal respected even where a relative has consented. Supply the surrogate form as a separate annexure.

What happens to positive screens. This is the point most often missing and the one that matters most. If you administer a depression instrument to two hundred older adults, some will screen positive and a few will endorse the item on self-harm or wishing to be dead. A protocol that collects those responses as data without saying what is done about them is not acceptable. Write the pathway down: who is informed, that the participant is offered referral to the psychiatry or geriatric service the same day, and that any indication of immediate risk is escalated to the treating consultant rather than recorded and set aside. State the equivalent pathway for participants found to have significant cognitive impairment, severe malnutrition or an unrecognised fall risk — each should be referred into routine care, and saying so strengthens the protocol rather than complicating it.

Caregivers are participants. In the caregiver burden topics the caregiver is a research subject with their own consent, their own information sheet and their own confidentiality — particularly since burden instruments ask questions a caregiver may not want the older person to see. State that responses are not shared with the patient or the family.

Consent that can actually be understood. Provide the information sheet in the local language, written simply and designed to be read aloud, with a provision for thumb impression and an impartial witness where the participant cannot read or sign. Allow time; consent obtained in a crowded corridor from someone who cannot hear the explanation is not consent.

Incidental disclosure. Studies asking about living arrangements, social support and loneliness occasionally surface neglect or mistreatment. State how such a disclosure is handled and to whom it is escalated.

Purely record-based studies may be granted a waiver of consent, applied for explicitly. Clearance commonly takes six to ten weeks and retrospective approval is not granted.

Work backwards from three questions. First, what has been used in comparable Indian studies? Choosing the instrument that dominates the local literature makes your findings comparable and gives you a prevalence estimate for the sample size calculation; choosing an unusual tool leaves your result floating with nothing to sit beside. Second, does a validated version exist in the language you will actually speak in the clinic? If not, either pick a different instrument or build the translation and validation into the study as a declared component. Third, can you administer it in the time available to every participant, including those who are hard of hearing, tired or accompanied by an anxious relative? A comprehensive battery that works beautifully in a research unit becomes unworkable when you are also running the clinic. Then check the practicalities before you commit: whether the instrument requires a licence, registration or written permission; whether it needs equipment such as a calibrated dynamometer or a measured walking course; whether its cut-offs were derived in a population resembling yours, which matters especially for cognitive instruments in low-literacy settings and for sarcopenia, where Asian criteria differ from European ones; and whether it can be administered by proxy for participants who cannot self-report. Write your answers into the methodology as a short paragraph per instrument. That paragraph is what separates a geriatrics protocol that survives review from one that comes back asking how frailty was defined.

Substantially. A PhD proposal typically runs 6,000 to 10,000 words and carries an extended critical literature review, a theoretical framework, a detailed methodology chapter and a discussion of expected contribution to the field. An MD synopsis is a two to four page registration document. The research question can be the same; the depth expected is not.

Most universities require registration within six to nine months of joining. Shortlist in the first two months, finalise with your guide by the third, and file for ethics clearance immediately afterwards. Because almost every topic here requires you to administer instruments in person, recruitment is slower than residents expect and cannot be accelerated at the end; secure instrument permissions and translations during the ethics review rather than after it, and close the collection window at least six months before submission.